Simple and Efficient Programming with C#: Skills to Build Applications with Visual Studio and .NET [1 ed.] 1484273214, 9781484273210

Apply skills and approaches to your programming to build a real-world application in C# 9 using the latest editions of V

780 211 5MB

English Pages 297 [282] Year 2021

Report DMCA / Copyright

DOWNLOAD FILE

Polecaj historie

Simple and Efficient Programming with C#: Skills to Build Applications with Visual Studio and .NET [1 ed.]
 1484273214, 9781484273210

Table of contents :
Table of Contents
About the Author
About the Technical Reviewer
Acknowledgments
Introduction
Part I: Fundamentals
Chapter 1: Flexible Code Using Polymorphism
Recap
Initial Program
Demonstration 1
Output
Analysis
Better Program
Demonstration 2
Analysis
Summary
Chapter 2: Abstract Class or Interface?
Recap
Initial Program
Better Program
Demonstration
Output
Analysis
Summary
Chapter 3: Wise Use of Code Comments
Recap
Initial Program
Demonstration 1
Output
Analysis
Better Program
Demonstration 2
Analysis
Use the Power of C#
Summary
Part II: Important Principles
Chapter 4: Know SOLID Principles
Single Responsibility Principle (SRP)
Initial Program
Demonstration 1
Output
Analysis
Better Program
Demonstration 2
Output
Open/Closed Principle (OCP)
Initial Program
Demonstration 3
Output
Analysis
Better Program
Demonstration 4
Output
Analysis
Liskov Substitution Principle (LSP)
Initial Program
Demonstration 5
Output
Better Program
Demonstration 6
Output
Analysis
Interface Segregation Principle (ISP)
Initial Program
Demonstration 7
Output
Analysis
Better Program
Demonstration 8
Output
Analysis
Dependency Inversion Principle (DIP)
Initial Program
Demonstration 9
Output
Analysis
Better Program
Demonstration 10
Output
Analysis
Summary
Chapter 5: Use the DRY Principle
Reasons for DRY
Initial Program
Demonstration 1
Output
Analysis
Better Program
Demonstration 2
Output
Analysis
Demonstration 3
Output
Demonstration 4
Output
Summary
Part III: Make Efficient Applications
Chapter 6: Separate Changeable Code Using Factories
The Problem Statement
Initial Program
Demonstration 1
Output
Analysis
Better Program
Demonstration 2
Output
Analysis
A New Requirement
Demonstration 3
Output
Analysis
Demonstration 4
Output
Analysis
Summary
Chapter 7: Add Features Using Wrappers
The Problem Statement
Using Subclassing
Using Object Composition
Class Diagram
Demonstration
Output
Analysis
Summary
Chapter 8: Efficient Templates Using Hooks
The Problem Statement
Initial Program
Class Diagram
Demonstration 1
Output
Analysis
Enhanced Requirement
Demonstration 2
Output
Summary
Chapter 9: Simplify Complex Systems Using Facades
The Problem Statement
Initial Program
Demonstration 1
Output
Analysis
Better Program
Class Diagram
Demonstration 2
Output
Analysis
Summary
Part IV: The Road Ahead
Chapter 10: Memory Management
Overview
Stack Memory vs. Heap Memory
Q&A Session
The Garbage Collector in Action
Different Phases of Garbage Collection
Different Cases of Invoking the Garbage Collector
Demonstration 1
Output
Analysis
Disposing of an Object
Finalize vs. Dispose
Demonstration 2
Output
Analysis
Q&A Session
Memory Leak Analysis
Demonstration 3
Snapshots from Diagnostic Tools
Summary
Chapter 11: Leftover Discussions
Static Method or Instance Method?
Recap
Learn Design Patterns
Brief History of Design Patterns
Here Is the Good News!
Q&A Session
Avoid Anti-patterns
Brief History of Anti-patterns
Examples of Anti-patterns
Types of Anti-patterns
Q&A Session
Some Common Terminology
Q&A Session
Summary
Appendix A: Winning Notes
A Personal Appeal to You
Appendix B: Resources
Index

Citation preview

Simple and Efficient Programming with C# Skills to Build Applications with Visual Studio and .NET — Vaskaran Sarcar

Simple and Efficient Programming with C# Skills to Build Applications with Visual Studio and .NET

Vaskaran Sarcar

Simple and Efficient Programming with C#: Skills to Build Applications with Visual Studio and .NET Vaskaran Sarcar Kolkata, West Bengal, India ISBN-13 (pbk): 978-1-4842-7321-0 https://doi.org/10.1007/978-1-4842-7322-7

ISBN-13 (electronic): 978-1-4842-7322-7

Copyright © 2021 by Vaskaran Sarcar This work is subject to copyright. All rights are reserved by the publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. Trademarked names, logos, and images may appear in this book. Rather than use a trademark symbol with every occurrence of a trademarked name, logo, or image we use the names, logos, and images only in an editorial fashion and to the benefit of the trademark owner, with no intention of infringement of the trademark. The use in this publication of trade names, trademarks, service marks, and similar terms, even if they are not identified as such, is not to be taken as an expression of opinion as to whether or not they are subject to proprietary rights. While the advice and information in this book are believed to be true and accurate at the date of publication, neither the authors nor the editors nor the publisher can accept any legal responsibility for any errors or omissions that may be made. The publisher makes no warranty, express or implied, with respect to the material contained herein. Managing Director, Apress Media LLC: Welmoed Spahr Acquisitions Editor: Smriti Srivastava Development Editor: Laura Berendson Coordinating Editor: Shrikant Vishwakarma Cover designed by eStudioCalamar Cover image designed by Pexels Distributed to the book trade worldwide by Springer Science+Business Media LLC, 1 New York Plaza, Suite 4600, New York, NY 10004. Phone 1-800-SPRINGER, fax (201) 348-4505, email [email protected] com, or visit www.springeronline.com. Apress Media, LLC is a California LLC, and the sole member (owner) is Springer Science+Business Media Finance Inc (SSBM Finance Inc). SSBM Finance Inc is a Delaware corporation. For information on translations, please e-mail [email protected]; for reprint, paperback, or audio rights, please e-mail [email protected] or visit http://www.apress. com/rights-permissions. Apress titles may be purchased in bulk for academic, corporate, or promotional use. eBook versions and licenses are also available for most titles. For more information, reference our Print and eBook Bulk Sales web page at http://www.apress.com/bulk-sales. Any source code or other supplementary material referenced by the author in this book is available to readers on GitHub via the book’s product page, located at www.apress.com/978-­1-­4842-­7321-­0. For more detailed information, please visit http://www.apress.com/source-­code. Printed on acid-free paper

Dear Reader, You inspire me with your loving comments. I get upset by your critical comments. But in every case, you help me grow into a better person and a better author. So, you are my teachers. I dedicate this work with love to you. I also try my best to help you grow.

Table of Contents About the Author��������������������������������������������������������������������������������������������������� xiii About the Technical Reviewer���������������������������������������������������������������������������������xv Acknowledgments�������������������������������������������������������������������������������������������������xvii Introduction������������������������������������������������������������������������������������������������������������xix

Part I: Fundamentals�������������������������������������������������������������������������������������� 1 Chapter 1: Flexible Code Using Polymorphism��������������������������������������������������������� 3 Recap�������������������������������������������������������������������������������������������������������������������������������������������� 3 Initial Program������������������������������������������������������������������������������������������������������������������������� 3 Demonstration 1���������������������������������������������������������������������������������������������������������������������� 4 Output�������������������������������������������������������������������������������������������������������������������������������������� 5 Analysis����������������������������������������������������������������������������������������������������������������������������������� 5 Better Program������������������������������������������������������������������������������������������������������������������������������ 7 Demonstration 2���������������������������������������������������������������������������������������������������������������������� 7 Analysis����������������������������������������������������������������������������������������������������������������������������������� 8 Summary������������������������������������������������������������������������������������������������������������������������������������ 14

Chapter 2: Abstract Class or Interface?����������������������������������������������������������������� 17 Recap������������������������������������������������������������������������������������������������������������������������������������������ 17 Initial Program���������������������������������������������������������������������������������������������������������������������������� 21 Better Program���������������������������������������������������������������������������������������������������������������������������� 25 Demonstration����������������������������������������������������������������������������������������������������������������������� 25 Output������������������������������������������������������������������������������������������������������������������������������������ 32 Analysis��������������������������������������������������������������������������������������������������������������������������������� 33 Summary������������������������������������������������������������������������������������������������������������������������������������ 34

v

Table of Contents

Chapter 3: Wise Use of Code Comments����������������������������������������������������������������� 35 Recap������������������������������������������������������������������������������������������������������������������������������������������ 35 Initial Program���������������������������������������������������������������������������������������������������������������������������� 37 Demonstration 1�������������������������������������������������������������������������������������������������������������������� 37 Output������������������������������������������������������������������������������������������������������������������������������������ 38 Analysis��������������������������������������������������������������������������������������������������������������������������������� 38 Better Program���������������������������������������������������������������������������������������������������������������������������� 39 Demonstration 2�������������������������������������������������������������������������������������������������������������������� 39 Analysis��������������������������������������������������������������������������������������������������������������������������������� 40 Use the Power of C#�������������������������������������������������������������������������������������������������������������������� 40 Summary������������������������������������������������������������������������������������������������������������������������������������ 43

Part II: Important Principles������������������������������������������������������������������������� 47 Chapter 4: Know SOLID Principles�������������������������������������������������������������������������� 49 Single Responsibility Principle (SRP)������������������������������������������������������������������������������������������ 50 Initial Program����������������������������������������������������������������������������������������������������������������������� 51 Demonstration 1�������������������������������������������������������������������������������������������������������������������� 51 Output������������������������������������������������������������������������������������������������������������������������������������ 53 Analysis��������������������������������������������������������������������������������������������������������������������������������� 54 Better Program���������������������������������������������������������������������������������������������������������������������� 54 Demonstration 2�������������������������������������������������������������������������������������������������������������������� 54 Output������������������������������������������������������������������������������������������������������������������������������������ 57 Open/Closed Principle (OCP)������������������������������������������������������������������������������������������������������� 58 Initial Program����������������������������������������������������������������������������������������������������������������������� 59 Demonstration 3�������������������������������������������������������������������������������������������������������������������� 61 Output������������������������������������������������������������������������������������������������������������������������������������ 64 Analysis��������������������������������������������������������������������������������������������������������������������������������� 65 Better Program���������������������������������������������������������������������������������������������������������������������� 66 Demonstration 4�������������������������������������������������������������������������������������������������������������������� 67

vi

Table of Contents

Output������������������������������������������������������������������������������������������������������������������������������������ 70 Analysis��������������������������������������������������������������������������������������������������������������������������������� 71 Liskov Substitution Principle (LSP)��������������������������������������������������������������������������������������������� 72 Initial Program����������������������������������������������������������������������������������������������������������������������� 77 Demonstration 5�������������������������������������������������������������������������������������������������������������������� 79 Output������������������������������������������������������������������������������������������������������������������������������������ 81 Better Program���������������������������������������������������������������������������������������������������������������������� 84 Demonstration 6�������������������������������������������������������������������������������������������������������������������� 84 Output������������������������������������������������������������������������������������������������������������������������������������ 88 Analysis��������������������������������������������������������������������������������������������������������������������������������� 88 Interface Segregation Principle (ISP)������������������������������������������������������������������������������������������ 88 Initial Program����������������������������������������������������������������������������������������������������������������������� 89 Demonstration 7�������������������������������������������������������������������������������������������������������������������� 93 Output������������������������������������������������������������������������������������������������������������������������������������ 94 Analysis��������������������������������������������������������������������������������������������������������������������������������� 94 Better Program���������������������������������������������������������������������������������������������������������������������� 95 Demonstration 8�������������������������������������������������������������������������������������������������������������������� 95 Output������������������������������������������������������������������������������������������������������������������������������������ 97 Analysis��������������������������������������������������������������������������������������������������������������������������������� 97 Dependency Inversion Principle (DIP)����������������������������������������������������������������������������������������� 98 Initial Program����������������������������������������������������������������������������������������������������������������������� 99 Demonstration 9������������������������������������������������������������������������������������������������������������������ 100 Output���������������������������������������������������������������������������������������������������������������������������������� 101 Analysis������������������������������������������������������������������������������������������������������������������������������� 102 Better Program�������������������������������������������������������������������������������������������������������������������� 102 Demonstration 10���������������������������������������������������������������������������������������������������������������� 103 Output���������������������������������������������������������������������������������������������������������������������������������� 105 Analysis������������������������������������������������������������������������������������������������������������������������������� 105 Summary���������������������������������������������������������������������������������������������������������������������������������� 106

vii

Table of Contents

Chapter 5: Use the DRY Principle�������������������������������������������������������������������������� 109 Reasons for DRY����������������������������������������������������������������������������������������������������������������������� 109 Initial Program�������������������������������������������������������������������������������������������������������������������������� 112 Demonstration 1������������������������������������������������������������������������������������������������������������������ 112 Output���������������������������������������������������������������������������������������������������������������������������������� 113 Analysis������������������������������������������������������������������������������������������������������������������������������� 114 Better Program�������������������������������������������������������������������������������������������������������������������������� 114 Demonstration 2������������������������������������������������������������������������������������������������������������������ 115 Output���������������������������������������������������������������������������������������������������������������������������������� 116 Analysis������������������������������������������������������������������������������������������������������������������������������� 117 Demonstration 3������������������������������������������������������������������������������������������������������������������ 118 Output���������������������������������������������������������������������������������������������������������������������������������� 121 Demonstration 4������������������������������������������������������������������������������������������������������������������ 123 Output���������������������������������������������������������������������������������������������������������������������������������� 127 Summary���������������������������������������������������������������������������������������������������������������������������������� 128

Part III: Make Efficient Applications����������������������������������������������������������� 129 Chapter 6: Separate Changeable Code Using Factories���������������������������������������� 131 The Problem Statement������������������������������������������������������������������������������������������������������������ 132 Initial Program�������������������������������������������������������������������������������������������������������������������������� 132 Demonstration 1������������������������������������������������������������������������������������������������������������������ 133 Output���������������������������������������������������������������������������������������������������������������������������������� 135 Analysis������������������������������������������������������������������������������������������������������������������������������� 135 Better Program�������������������������������������������������������������������������������������������������������������������������� 136 Demonstration 2������������������������������������������������������������������������������������������������������������������ 138 Output���������������������������������������������������������������������������������������������������������������������������������� 140 Analysis������������������������������������������������������������������������������������������������������������������������������� 140 A New Requirement������������������������������������������������������������������������������������������������������������������ 141 Demonstration 3������������������������������������������������������������������������������������������������������������������ 141 Output���������������������������������������������������������������������������������������������������������������������������������� 143 Analysis������������������������������������������������������������������������������������������������������������������������������� 144 viii

Table of Contents

Demonstration 4������������������������������������������������������������������������������������������������������������������ 144 Output���������������������������������������������������������������������������������������������������������������������������������� 147 Analysis������������������������������������������������������������������������������������������������������������������������������� 147 Summary���������������������������������������������������������������������������������������������������������������������������������� 148

Chapter 7: Add Features Using Wrappers������������������������������������������������������������� 149 The Problem Statement������������������������������������������������������������������������������������������������������������ 150 Using Subclassing��������������������������������������������������������������������������������������������������������������������� 150 Using Object Composition��������������������������������������������������������������������������������������������������������� 153 Class Diagram��������������������������������������������������������������������������������������������������������������������� 158 Demonstration��������������������������������������������������������������������������������������������������������������������� 159 Output���������������������������������������������������������������������������������������������������������������������������������� 163 Analysis������������������������������������������������������������������������������������������������������������������������������� 165 Summary���������������������������������������������������������������������������������������������������������������������������������� 166

Chapter 8: Efficient Templates Using Hooks��������������������������������������������������������� 169 The Problem Statement������������������������������������������������������������������������������������������������������������ 169 Initial Program�������������������������������������������������������������������������������������������������������������������������� 170 Class Diagram��������������������������������������������������������������������������������������������������������������������� 173 Demonstration 1������������������������������������������������������������������������������������������������������������������ 173 Output���������������������������������������������������������������������������������������������������������������������������������� 176 Analysis������������������������������������������������������������������������������������������������������������������������������� 176 Enhanced Requirement������������������������������������������������������������������������������������������������������������� 177 Demonstration 2������������������������������������������������������������������������������������������������������������������ 181 Output���������������������������������������������������������������������������������������������������������������������������������� 184 Summary���������������������������������������������������������������������������������������������������������������������������������� 184

Chapter 9: Simplify Complex Systems Using Facades����������������������������������������� 187 The Problem Statement������������������������������������������������������������������������������������������������������������ 188 Initial Program�������������������������������������������������������������������������������������������������������������������������� 189 Demonstration 1������������������������������������������������������������������������������������������������������������������ 191 Output���������������������������������������������������������������������������������������������������������������������������������� 192 Analysis������������������������������������������������������������������������������������������������������������������������������� 193 ix

Table of Contents

Better Program�������������������������������������������������������������������������������������������������������������������������� 193 Class Diagram��������������������������������������������������������������������������������������������������������������������� 193 Demonstration 2������������������������������������������������������������������������������������������������������������������ 194 Output���������������������������������������������������������������������������������������������������������������������������������� 197 Analysis������������������������������������������������������������������������������������������������������������������������������� 198 Summary���������������������������������������������������������������������������������������������������������������������������������� 198

Part IV: The Road Ahead����������������������������������������������������������������������������� 201 Chapter 10: Memory Management����������������������������������������������������������������������� 203 Overview����������������������������������������������������������������������������������������������������������������������������������� 203 Stack Memory vs. Heap Memory���������������������������������������������������������������������������������������������� 205 Q&A Session������������������������������������������������������������������������������������������������������������������������ 208 The Garbage Collector in Action������������������������������������������������������������������������������������������������ 212 Different Phases of Garbage Collection������������������������������������������������������������������������������� 213 Different Cases of Invoking the Garbage Collector�������������������������������������������������������������� 213 Demonstration 1������������������������������������������������������������������������������������������������������������������ 216 Output���������������������������������������������������������������������������������������������������������������������������������� 218 Analysis������������������������������������������������������������������������������������������������������������������������������� 219 Disposing of an Object�������������������������������������������������������������������������������������������������������������� 220 Finalize vs. Dispose������������������������������������������������������������������������������������������������������������������ 221 Demonstration 2������������������������������������������������������������������������������������������������������������������ 224 Output���������������������������������������������������������������������������������������������������������������������������������� 226 Analysis������������������������������������������������������������������������������������������������������������������������������� 226 Memory Leak Analysis�������������������������������������������������������������������������������������������������������������� 231 Demonstration 3������������������������������������������������������������������������������������������������������������������ 233 Snapshots from Diagnostic Tools���������������������������������������������������������������������������������������� 236 Summary���������������������������������������������������������������������������������������������������������������������������������� 238

x

Table of Contents

Chapter 11: Leftover Discussions������������������������������������������������������������������������� 241 Static Method or Instance Method?������������������������������������������������������������������������������������������ 241 Recap���������������������������������������������������������������������������������������������������������������������������������� 241 Learn Design Patterns�������������������������������������������������������������������������������������������������������������� 244 Brief History of Design Patterns������������������������������������������������������������������������������������������ 245 Here Is the Good News!������������������������������������������������������������������������������������������������������� 248 Q&A Session������������������������������������������������������������������������������������������������������������������������ 249 Avoid Anti-patterns������������������������������������������������������������������������������������������������������������������� 250 Brief History of Anti-patterns����������������������������������������������������������������������������������������������� 251 Examples of Anti-patterns��������������������������������������������������������������������������������������������������� 252 Types of Anti-patterns��������������������������������������������������������������������������������������������������������� 253 Q&A Session������������������������������������������������������������������������������������������������������������������������ 254 Some Common Terminology����������������������������������������������������������������������������������������������������� 257 Q&A Session������������������������������������������������������������������������������������������������������������������������ 259 Summary���������������������������������������������������������������������������������������������������������������������������������� 261

Appendix A: Winning Notes���������������������������������������������������������������������������������� 263 A Personal Appeal to You���������������������������������������������������������������������������������������������������������� 263

Appendix B: Resources����������������������������������������������������������������������������������������� 265 Index��������������������������������������������������������������������������������������������������������������������� 267

xi

About the Author Vaskaran Sarcar obtained his Master of Engineering in software engineering from Jadavpur University, Kolkata, India, and an MCA from Vidyasagar University, Midnapore, India. He was a National Gate Scholar (2007–2009) and has more than 12 years of experience in education and the IT industry.  Vaskaran devoted his early years (2005–2007) to the teaching profession at various engineering colleges, and later he joined HP India PPS R&D Hub Bangalore. He worked there until August 2019. At the time of his retirement from HP, he was a senior software engineer and team lead. To follow his dream and passion, Vaskaran is now an independent, full-time author. His other Apress books include the following: •

Design Patterns in C# Second Edition (Apress, 2020)



Getting Started with Advanced C# (Apress, 2020)



Interactive Object-Oriented Programming in Java Second Edition (Apress, 2019)



Java Design Patterns Second Edition (Apress, 2019)



Design Patterns in C# (Apress, 2018)



Interactive C# (Apress, 2017)



Interactive Object-Oriented Programming in Java (Apress, 2016)



Java Design Patterns (Apress, 2016)

The following list includes his non-Apress books: •

Python Bookcamp (Amazon, 2021)



Operating System: Computer Science Interview Series (Createspace, 2014) xiii

About the Technical Reviewer Carsten Thomsen is primarily a back-end developer, but works with smaller front-end bits as well. He has authored and reviewed a number of books and created numerous Microsoft Learning courses, all to do with software development. He works as a freelancer/contractor in various countries in Europe; Azure, Visual Studio, Azure DevOps, and GitHub are some of the tools he works with. Being an exceptional troubleshooter—asking the right questions, including the less logical ones, in a most-­logical to leastlogical fashion—he also enjoys working with architecture, research, analysis, development, testing, and bug fixing. Carsten is a very good communicator with great mentoring and team-lead skills, and fantastic skills in researching and presenting new material.  

xv

Acknowledgments First, I thank the Almighty. I sincerely believe that with HIS blessings only could I complete this book. I also extend my deepest gratitude and thanks to the following: Ratanlal Sarkar and Manikuntala Sarkar: My dear parents, thanks for all your support towards me. Indrani, my wife; Ambika, my daughter; Aryaman, my son: Sweethearts, I love you all. Sambaran, my brother: Thank you for your constant encouragement toward me. Carsten: You are a great technical advisor. Whenever I was in need, your support was there. Thank you one more time. Celestin, Laura, and Smriti: Thanks for giving me another opportunity to work with you and Apress. Shrikant, Nirmal, Sherly, Sankar and Mohan: Thank you for your exceptional support to finalize my work. Your efforts are extraordinary.

xvii

Introduction Welcome to your journey through Simple and Efficient Programming with C#: Skills to Build Applications with Visual Studio and .NET. C# is an object-oriented programming (OOP) language. You may already know C# keywords, or even some interesting features. You may also know how to write simple programs in C#. You can learn these things from an introductory book or an online tutorial. These are useful things to know, but they are not sufficient to understand an enterprise codebase. This is why a novice programmer often finds it difficult to understand an expert’s code. He or she wonders why an experienced programmer wrote the program differently. It may appear to the novice that the expert could have used an easier approach to solve the problem. But there are reasons why an experienced programmer might follow a different approach. The word “experienced” indicates that these programmers have more experience in programming and know the pros and cons of different approaches. They know how the C# features can be used in the best possible way to develop an application. So, the applications they make are usually powerful. What do I mean by a powerful application? For me, a powerful application is robust, extensible, and easily maintainable, but simple to use. This book is an introductory guide to develop such applications. This is the core aim of this book. To write better quality programs, senior programmers follow in experts’ footprints. They learn from collective wisdom and recorded experience from the past. So, instead of attempting an entirely new solution, you should first consider this knowledge base, which will help you produce better quality code. It is best to have some idea about why you should or shouldn’t follow any specific guideline. Malcolm Gladwell, in his book Outliers (Little, Brown and Company), discussed the 10,000-hour rule. This rule says that the key to achieving world-class expertise in any skill is, to a large extent, a matter of practicing the correct way, for a total of around 10,000 hours. I acknowledge that it is impossible to consider all experiences before you write a program. Also, sometimes it is OK to bend the rules if the return on investment (ROI) is nice. So, keep in mind the Pareto principle, or 80-20 rule. This rule simply states that 80% of outcomes come from 20% of all causes. This is useful in programming too. When you identify the most essential characteristics of top-quality programs and use xix

Introduction

them in your applications, you also qualify yourself as an experienced programmer, and your application will be robust, flexible, and maintainable. In this book, I share with you these important principles, which will help you write better programs for case studies. Some of these principles you may know already, but when you see them in action and compare the case studies, you’ll understand their importance.

How Is the Book Organized? The book has four major parts, which are as follows:

xx



The first three chapters form Part I, in which there is a detailed discussion of polymorphism and the use of abstract classes and interfaces. Here, code comments will be examined, and you will learn when to use them effectively. These are the fundamental building blocks for the rest of the book.



In the world of programming, there is no shortage of programming principles and design guidelines. Each of these suggestions has its own benefits. To become a professional programmer, you do not need to learn everything at the same time. So, in Part II, I discuss six design principles, which include SOLID principles and the DRY principle. These are the foundation of well-known design patterns. Once you understand them, you can consider yourself a better programmer.



The best way of learning is by doing and analyzing case studies. So, in Part III of the book, you will see interesting applications that use some well-known patterns. This part gives you hints about how a professional coder develops an enterprise application.



There is no end to learning. So, Part IV includes some interesting topics such as how to prevent memory leaks, how to choose between a static method and an instance method, and some common terms from software development that are not discussed in detail in this book. A quick overview of these topics will help you to be familiar with them when you see them in your future endeavors.

Introduction



You can download all the source code for the book from the publisher’s website. I have a plan to maintain the “errata,” and, if required, I can also make some updates/announcements there. So, I suggest that you visit those pages to receive any important corrections or updates.

Prerequisite Knowledge This book is intended for those who are familiar with the basic language constructs of C# and have an idea about pure object-oriented concepts like polymorphism, inheritance, abstraction, encapsulation, and, most important, how to compile or run a C# application in Visual Studio. This book does not invest time in easily available topics, such as how to install Visual Studio on your system, or how to write a “Hello World” program in C#, or how you can use an if-else statement or a while loop, etc. This book is written using the most basic features of C# so that for most of the programs herein you do not need to be familiar with advanced topics in C#. The examples are simple and straightforward. I believe that they are written in such a way that even if you are familiar with another popular language such as Java, C++, and so on, you can still easily grasp the concepts in this book.

Who Is This Book For? In short, you can pick up this book if the answer is “yes” to the following questions: •

Are you familiar with basic constructs in C# and object-oriented concepts like polymorphism, inheritance, abstraction, and encapsulation?



Do you know how to set up your coding environment?



Have you completed at least one basic course on C# and now are interested in writing better programs? Are you also interested to know how a professional programmer designs his or her applications?



Are you interested in knowing how the core constructs of C# work behind standard design patterns?

xxi

Introduction

You probably shouldn’t pick this book if the answer is yes to any of the following questions: •

Are you absolutely new to C#?



Are you looking for advanced concepts in C#, excluding the topics mentioned previously?



Are you interested in exploring a book where the focus is not on standard design principles?



“I do not like Windows, Visual Studio, and/or .NET. I want to learn and use C# without them only.” —Is this statement true for you?

Guidelines for Using This Book To use this book more effectively, consider the following:

xxii



This book works best if you’ve gone through an introductory course on C# and are familiar with the common terms, such as polymorphism, and have heard about abstract classes and interfaces. If this is not the case, please read about these topics before you start reading this book.



I suggest you go through the chapters sequentially. This is because some fundamental design techniques may have been discussed in a previous chapter and I have not repeated those techniques in later chapters.



I started this book using Microsoft Visual Studio Community 2019 (Version 16.8.4) in a Windows 10 environment. This community edition is free of charge. If you do not use the Windows operating system, you can use Visual Studio Code, which is also a source-­ code editor developed by Microsoft to support Windows, Linux, or Mac operating systems. This multi-platform IDE is also free. When I started the book, I started with the latest versions of C# that were available at that time. In this context, it is useful to know that

Introduction

nowadays the C# language version is automatically selected based on your project’s target framework(s), so you can always get the highest compatible version by default. In the latest versions, Visual Studio doesn’t support changing the version value in the user interface, but you can change it by editing the csproj file. •

Later, I also used Microsoft Visual Studio Community 2019 Preview 4.0 and set my target framework to .NET 6.0. As per the new rule, you can simply say that when your target framework is .NET 5.x (and later), you’ll get C# 9.0 and later by default. If you are interested in the C# language versioning, you can go to this link: https://docs. microsoft.com/en-­us/dotnet/csharp/language-­reference/ configure-­language-­version.



Version updates will come continuously, but I strongly believe that these version details should not matter much to you because I have used the fundamental constructs of C#. So, the code in this book should execute smoothly in the upcoming versions of C#/Visual Studio as well. Though I also believe that the results should not vary in other environments, you know the nature of software—it is naughty. So, I recommend that if you want to see the exact same output, you mimic the same environment.



You can download and install the Visual Studio IDE from https:// visualstudio.microsoft.com/downloads/. You are expected to see the screen shown in Figure 1.

xxiii

Introduction

Figure 1.  Download link for Visual Studio 2019 and Visual Studio Code

Note At the time of writing, this link works fine, and the information is correct. But the link and policies may change in the future. •

I have installed the class designer component in Visual Studio 2019 to draw class diagrams for my programs. But I needed to edit some of these diagrams for better readability. For example, I added some valuable notes in some diagrams, so that you can understand them easily.

Conventions Used in This Book Here, I will mention only two points: In many places, to avoid more typing, I have used the word “his” only. Please treat it as “his” or “her,” whichever is applicable for you. Second, all the outputs and codes of the book follow the same font and structure. To draw your attention, in some places, I have made them bold. For example, consider the following code fragment (taken from Chapter 4 when I discuss LSP) and the lines in bold. xxiv

Introduction

... // Instantiating two registered users RegisteredUser robin = new RegisteredUser("Robin"); RegisteredUser jack = new RegisteredUser("Jack"); // Adding the users to usermanager helper.AddUser(robin); helper.AddUser(jack); GuestUser guestUser1 = new GuestUser(); helper.AddUser(guestUser1); // Processing the payments using // the helper class. // You can see the problem now. ....

Final Words I must say that you are an intelligent person. You have chosen a subject that can assist you throughout your career. If you are a developer/programmer, you need these concepts. If you are an architect of a software organization, you need these concepts. If you are a college student, you need these concepts, not only to score well on exams but also to enter the corporate world. Even if you are a tester who needs to take care of white-­ box testing or simply needs to know about the code paths of a product, these concepts will help you a lot. Remember that you have just started on this journey. As you learn about these concepts, I suggest you write your own code; only then will you master this area. There is no shortcut for this. Do you remember Euclid’s reply to the ruler? If not, let me remind you of his reply: There is no royal road to geometry. So, study and code; understand a new concept and code again. Do not give up when you face challenges. They are the indicators that you are growing better. I believe that this book is designed for you in such a way that upon its completion, you will have developed an adequate knowledge of the topic, and, most important, you’ll know how to go further. Lastly, I hope that this book can provide help to you and that you will value the effort. xxv

PART I

Fundamentals Part I consists of three chapters, in which we will discuss the following questions: •

How can we use the power of polymorphism and why is it beneficial?



How can we combine an abstract class and interfaces to make an efficient application?



How can we use meaningful code comments and avoid unnecessary comments in a program?

Almost every C# application uses comments, the concept of polymorphism, and abstract classes and interfaces. When we implement these techniques in a better way, the program is better. I consider them the fundamental techniques for an efficient application.

CHAPTER 1

Flexible Code Using Polymorphism Ask a developer, “What are the fundamental characteristics of object-oriented programming (OOP)?” and you will get an immediate reply saying, “Classes (and objects), inheritance, abstraction, encapsulation, and polymorphism are the most important characteristics in OOP”. In addition, when you analyze enterprise code that is based on OOP, you’ll find different forms of polymorphism. But the truth is, a novice programmer rarely uses the power of polymorphism. This chapter focuses on this topic. It shows you some simple but powerful code examples using this principle.

R  ecap Polymorphism simply means there is one name with many forms. Consider the behavior of your pet dog. When it sees an unknown person, it starts barking. But when it sees you, it makes different noises and behaves differently. In both cases, this dog sees with his eyes but based on his observations he behaves differently. Polymorphic code can work in the same way. Consider a method that you might use to add some operands. If the operands were integers, you would expect to get a sum of the integers. But if you were to deal with string operands, you would expect to get a concatenated string.

I nitial Program Let’s look at a program that compiles and runs successfully. In this program, there are three different types of animals: tigers, dogs, and monkeys. Each of them can produce a different sound. So, there are classes with these names, and in each class, there is a Sound() method. See whether you can improve this program.

© Vaskaran Sarcar 2021 V. Sarcar, Simple and Efficient Programming with C#, https://doi.org/10.1007/978-1-4842-7322-7_1

3

Chapter 1

Flexible Code Using Polymorphism

Demonstration 1 Here is a program that does not use the concept of polymorphism. using System; namespace DemoWithoutPolymorphism {     class Tiger     {         public void Sound()         {             Console.WriteLine("Tigers roar.");         }     }     class Dog     {         public void Sound()         {             Console.WriteLine("Dogs bark.");         }     }     class Monkey     {         public void Sound()         {             Console.WriteLine("Monkeys whoop.");         }     }     class Program     {         static void Main(string[] args)         {             Console.WriteLine("***Sounds of the different animals.***");             Tiger tiger = new Tiger();             tiger.Sound(); 4

Chapter 1

Flexible Code Using Polymorphism

            Dog dog = new Dog();             dog.Sound();             Monkey monkey = new Monkey();             monkey.Sound();             Console.ReadKey();         }     } }

Output ***Sounds of the different animals.*** Tigers roar. Dogs bark. Monkeys whoop.

Analysis When you use Tiger tiger = new Tiger(); the tiger is a reference to an object that is based on the Tiger class. This reference refers to the object but does not contain the object data itself. Even Tiger tiger; is also a valid line of code that tells you to create an object reference without creating the object. Understand that when you use Tiger tiger = new Tiger(); you are programming to an implementation. Notice that in this case the reference and object are both of the same types. You can improve this program using the concept of polymorphism. In the upcoming implementation, I show you such an example. I use an interface in this example. I can achieve the same thing using an abstract class too. Before I show you the example, let me remind you of few important points: •

When you use an abstract class or an interface, the first thing that comes to mind is inheritance. How do you know whether you have correctly used inheritance? The simple answer is: you do an IS-A test. For example, a rectangle IS-A shape, but the reverse is not necessarily true. Take another example: a monkey IS-An animal but not all animals are monkeys. Notice that the IS-A test is unidirectional.

5

Chapter 1

Flexible Code Using Polymorphism



In programming, if you inherit class B from class A, you say that B is the subclass and A is the parent class or base class. But most important, you can say B is a type of A. So, if you derive a Tiger class or a Dog class from a base class called Animal (or an interface, say, IAnimal), you can say that Dog IS-An Animal (or IAnimal) or Tiger IS-An Animal (or IAnimal).



If you have an inheritance tree, this IS-A test can be applied anywhere in the tree. For example, a rectangle IS-A special type of shape. A square IS-A special type of rectangle. So, a square IS-A shape too.



Let us say we represent rectangles and shapes using the Rectangle and Shape classes, respectively. Now when we say Rectangle IS-A Shape, programmatically we mean a Rectangle instance can invoke the methods that a Shape instance can invoke. If needed, a Rectangle instance can invoke some additional methods too. These additional methods can be defined in the Rectangle class.

You know that a superclass reference can refer to a subclass object. Here you see that each tiger, dog, or monkey is an animal. So, you can introduce a supertype and inherit all these concrete classes from it. Let’s name the supertype IAnimal. Here is a code fragment that shows the IAnimal interface. It also gives you the idea of how you can override its Sound() method in the Tiger class. The Monkey and Dog classes can do the same thing. interface IAnimal {     void Sound(); } class Tiger : IAnimal {     public void Sound()     {         Console.WriteLine("Tigers roar.");     } }

6

Chapter 1

Flexible Code Using Polymorphism

Programming to a supertype gives you more flexibility. It allows you to use a reference variable polymorphically. The following code segment demonstrates such a usage: IAnimal animal = new Tiger(); animal.Sound(); animal = new Dog(); animal.Sound(); //remaining code skipped

Better Program I have rewritten this program, which produces the same output. Let’s have a look at the following demonstration.

Demonstration 2 This is a modified version of Demonstration 1. using System; namespace UsingPolymorphism {     interface IAnimal     {         void Sound();     }     class Tiger: IAnimal     {         public void Sound()         {             Console.WriteLine("Tigers roar.");         }     }

7

Chapter 1

Flexible Code Using Polymorphism

    class Dog: IAnimal     {         public void Sound()         {             Console.WriteLine("Dogs bark.");         }     }     class Monkey: IAnimal     {         public void Sound()         {             Console.WriteLine("Monkeys whoop.");         }     }     class Program     {         static void Main(string[] args)         {             Console.WriteLine("***Sounds of the different animals.***");             IAnimal animal = new Tiger();             animal.Sound();             animal = new Dog();             animal.Sound();             animal = new Monkey();             animal.Sound();             Console.ReadKey();         }     } }

Analysis Have you noticed the difference? Inside the Main() method, you use the superclass reference animal to refer to different derived class objects. 8

Chapter 1

Flexible Code Using Polymorphism

Now you not only type less, but you also use a program that is more flexible and easier to maintain. If you want, you could also iterate over a list. For example, you could replace the following code segment inside Main(): IAnimal animal = new Tiger(); animal.Sound(); animal = new Dog(); animal.Sound(); animal = new Monkey(); animal.Sound(); with the following code: List animals = new List {     new Tiger(),     new Dog(),     new Monkey() }; foreach (IAnimal animal in animals) animal.Sound(); If you run the program again with these changes, you will see the same output.

POINT TO REMEMBER When you use List, do not forget to include the following namespace at the beginning of the program: using System.Collections.Generic;

This discussion is not over yet. Here, I have used one of the simplest forms of polymorphism. In this case, a thought may come into your mind: we know a supertype reference can refer to a subtype object in C#. So, when I use the following lines: IAnimal animal = new Tiger(); animal.Sound(); 9

Chapter 1

Flexible Code Using Polymorphism

you can surely predict that the Sound() method of the Tiger class will be invoked. So, it appears that you know the output in advance, and you doubt the usefulness of the concept of polymorphism. If this is the case, consider the following discussion. Let us assume that you create a subtype based on some run-time conditions, such as a random number or a user input. In any of these cases, you cannot predict the output in advance. For example, see the following lines of code: IAnimal animal = GetAnimal(); animal.Sound(); What is the difference? Anyone who sees this code segment can assume that GetAnimal() returns an animal that can make a sound. How can you achieve this? It is pretty simple: let me rewrite the client code. Notice the changes in bold : class Program { static void Main() {   Console.WriteLine("***Sounds of the different animals.***");    IAnimal animal = GetAnimal();    animal.Sound();    animal = GetAnimal();    animal.Sound();    animal = GetAnimal();    animal.Sound();    Console.ReadKey(); } private static IAnimal GetAnimal() {   IAnimal animal;   Random random = new Random();   // Get a number between 0 and 3(exclusive)   int temp = random.Next(0, 3);

10

Chapter 1

Flexible Code Using Polymorphism

  if (temp == 0)   {    animal = new Tiger();   }   else if (temp == 1)   {    animal = new Dog();   }   else   {    animal = new Monkey();   }   return animal;   } } Run this application now and notice the output. Here is the sample output that I got on the various runs: First Run: ***Sounds of the different animals.*** Monkeys whoop. Dogs bark. Monkeys whoop. Second Run: ***Sounds of the different animals.*** Dogs bark. Dogs bark. Tigers roar. Third Run: ***Sounds of the different animals.*** Tigers roar. Monkeys whoop. Dogs bark. 11

Chapter 1

Flexible Code Using Polymorphism

Note  When you download the source code from the Apress website, refer to the folder PolymorphismDemo2 inside Chapter1 to see the complete program. It is now clear that no one can predict the output of this program in advance. You can see the power of polymorphism in this example. I’ll finish the chapter with a few more important points that will help you understand and use polymorphic code. You can replace animal.Sound(); with the following code: MakeSound(animal); Where MakeSound() is defined as follows: private static void MakeSound(IAnimal animal) {   animal.Sound(); } Why do I show this to you? By following this approach, you can pass a supertype reference to this method to invoke the appropriate subtype method. This gives you flexibility and helps you write more-readable code. Here is an alternative version of the client code that we just discussed: class Program     {         static void Main()         {             Console.WriteLine("***Sounds of the different animals.***");             IAnimal animal = GetAnimal();             MakeSound(animal);             animal = GetAnimal();             MakeSound(animal);             animal = GetAnimal();             MakeSound(animal);             Console.ReadKey();         }

12

Chapter 1

Flexible Code Using Polymorphism

        private static void MakeSound(IAnimal animal)         {             animal.Sound();         }         private static IAnimal GetAnimal()         {             IAnimal animal;             Random random = new Random();             // Get a number between 0 and 3(exclusive)             int temp = random.Next(0, 3);             if (temp == 0)             {                 animal = new Tiger();             }             else if (temp == 1)             {                 animal = new Dog();             }             else             {                 animal = new Monkey();             }             return animal;         }     } }

Note You should not assume that the GetAnimal() and MakeSound(...) methods need to be static only. You can use them as instance methods too. When you download the source code from the Apress website, refer to the folder PolymorphismDemo3 inside Chapter1 to see this modified program.

13

Chapter 1

Flexible Code Using Polymorphism

S  ummary To implement a polymorphic behavior, I started with an interface. I could achieve the same thing using an abstract class too. There are situations where an interface is a better choice than an abstract class, and vice versa. You will see a discussion of this in Chapter 2. When you code to a supertype (it can be an interface, an abstract class, or simply a parent class), the code can work with any new classes implementing the interface. This helps you to respond to changes in the future and adopt the updated requirements easily. This is the power of polymorphism. But if you use only concrete classes in your program, it is very likely that you will need to change your existing code in the future, such as when you add a new concrete class. This approach does not follow the Open/ Closed principle, which says your code should be open for extension but closed for modification. I have shown you the advantages of polymorphism. But it’s not always easy to write polymorphic code, and you need to be careful when you use it. You’ll get a better idea about this when I discuss SOLID principles in Chapter 4. Not everything in this chapter may be new to you, but I believe that you have a better grasp of polymorphism now. Before you move into the next chapters, let me make sure that we agree on these points and you are familiar with the following terminologies: When you write: Tiger tiger = new Tiger(); tiger.Sound(); you are programming to concrete implementation. When you write: IAnimal animal = new Tiger(); animal.Sound(); you are programming to a supertype. It is often referred to as programming to an interface.

Note  When we say “programming to an interface,” it does not necessarily mean that you use a C# interface only. It can be an abstract class, or a parent/ base class too.

14

Chapter 1

Flexible Code Using Polymorphism

You can follow a better approach when you write something like: IAnimal animal = GetAnimal(); animal.Sound(); In this case, no one can predict the output in advance merely by reading the code. In simple terms, this code segment implies that you announce to the outside world that you get an animal through the GetAnimal() method and this animal can make a sound. In short, this chapter answered the following questions: •

How can you perform an IS-A test?



How can you write a polymorphic code for your application and why is it better?



How can you iterate over a list when you write the polymorphic code?



How can you write a better polymorphic code?



How do experts differentiate between “programming to an implementation” and “programming to an interface”?

15

CHAPTER 2

Abstract Class or Interface? There are many code segments in which you can use an abstract class instead of a C# interface, and vice versa. If the code is small and is used to perform a simple task, you might not see the difference between these two techniques. However, when the code is large and extendable, the choice between them can play a vital role in performance and maintenance. In this chapter, we do not focus primarily on the basic differences between an abstract class and an interface. Instead, we discuss code segments where you can use either method, and the compiler does not raise any issues. Then, we will analyze how to write an efficient program combining both these techniques in some specific scenarios.

R  ecap Employers often ask job candidates to explain the differences between an abstract class and an interface. It is a common question, and I expect that you know the answer. Before I answer this and analyze these two important topics, let me remind you of some fundamental points so as to avoid confusion in the future: •

Normally an abstract class suits best when you share a common behavior across subclasses, but you want to guarantee that nobody can make an object from the class.



Interfaces are best when you define “roles” that other classes play by, and it does not matter whether these classes belong to the same inheritance tree or not. What does this mean? See the following discussion.



In Figure 2-1, you can see the Tiger and Dog classes inherit from the abstract class Animal. There is a Sound() method in these classes.

© Vaskaran Sarcar 2021 V. Sarcar, Simple and Efficient Programming with C#, https://doi.org/10.1007/978-1-4842-7322-7_2

17

Chapter 2

Abstract Class or Interface?

Figure 2-1.  Animal hierarchy •

In Figure 2-2, you see the TigerToy class and JumpingDog class inherit from the SoftToys class. Each class in this inheritance hierarchy contains a Sound() method too.

Figure 2-2.  SoftToys hierarchy

18

Chapter 2

Abstract Class or Interface?



Now tell me, though all of the Tiger, Dog, TigerToy, and JumpingDog instances can make sound, should you mix them? Or, can you say soft toys are animals or animals are soft toys? No. It is because the animal hierarchy and soft toy hierarchy are different. You should not treat a jumping dog (which is a soft toy) as a living animal just because it can make a sound.



But an interface can fit in this scenario. If you start with an interface, say, ISound, the Tiger class, Dog class, TigerToy class, and JumpingDog class can implement the interface and override the Sound() method as needed.



An abstract class has its own power and uses. For example, it can contain fields and concrete methods that an interface cannot contain. From C# 8.0 onward, you can include default methods. But normally an interface is like an abstract class that includes all the abstract methods.



In short, when you need to simulate behaviors from multiple classes, an interface is the right choice. This is because C# does not support the concept of multiple inheritances through classes.

DIAMOND PROBLEM Contemplate the following scenario: suppose there is an inheritance hierarchy in which the Shape class is placed on top. This class has a method called AboutMe(). Two classes, Triangle and Rectangle, derive from Shape. Both derived classes have redefined the AboutMe() method (in programming terms, they have overridden the method for their own purpose). The code may look like this: class Shape     {         public virtual void AboutMe()         {             Console.WriteLine("It is an arbitrary Shape.");         }     }

19

Chapter 2

Abstract Class or Interface?

    class Triangle : Shape     {         public override void AboutMe()         {             Console.WriteLine("It is a Triangle.");         }     }     class Rectangle : Shape     {         public override void AboutMe()         {             Console.WriteLine("It is a Rectangle");         }     }

Now, say a new class named GrandShape derives from both Triangle and Rectangle. Figure 2-3 shows a sample class diagram for this.

Figure 2-3.  Diamond problem resulting from multiple inheritances

20

Chapter 2

Abstract Class or Interface?

Now we have an ambiguity: from which class will GrandShape inherit or call AboutMe()? Is it from Triangle or is it from Rectangle? To remove this type of ambiguity, C# does not support the concept of multiple inheritances through the class. This problem is known by a famous name: the diamond problem. So, if you see code like this: class GrandShape: Triangle, Rectangle // Error: Diamond Effect {        // Some  code   }

you will notice that the C# compiler shows you the following error: CS1721   Class 'GrandShape' cannot have multiple base classes: 'Triangle' and 'Rectangle'

People often ask why C++ supports multiple inheritances. The same problem can exist there too. To answer this, I can share my personal opinion: C# designers wanted to avoid any unwanted outcomes in an application resulting from this kind of feature. Their key goal is to make the language simple and less error-prone. When you support special scenarios like this, you need to implement additional rules to validate them. Maintaining this kind of additional rule can make a programming language complex. In the end, it is up to the team that designs the language.

Initial Program Let us consider some vehicles that can either float or fly. Since boats float and airplanes fly, I use a Boat class and an Airplane class in the upcoming example. Since both are vehicles, you can start with an interface IVehicle to form the following inheritance hierarchy: interface IVehicle {     void Fly();     void Float(); }

21

Chapter 2

Abstract Class or Interface?

class Boat: IVehicle {     public void Float()     {         Console.WriteLine("I like to float.");     }     public void Fly()     {         throw new NotImplementedException();     } } class Airplane: IVehicle {     public void Float()     {         throw new NotImplementedException();     }     public void Fly()     {         Console.WriteLine("I like to fly.");     } } But you may prefer an abstract class over an interface. So, you may redesign the code as follows: abstract class Vehicle {     public abstract void Float();     public abstract void Fly(); } class Boat : Vehicle {     public override void Float()     { 22

Chapter 2

Abstract Class or Interface?

        Console.WriteLine("I like to float.");     }     public override void Fly()     {         throw new NotImplementedException();     } } class Airplane : Vehicle {     public override void Float()     {         throw new NotImplementedException();     }     public override void Fly()     {         Console.WriteLine("I like to fly.");     } } At this point, these two designs may appear to be the same. Now, let’s say you need to consider a new type of vehicle: ships. You know that if you consider only vehicles that either float or fly, you can put a common behavior inside the abstract class. In our case, out of these three types of vehicles, ships and boats float, but they do not fly. So, it may appear to you that you can make a common Float() method and move it to the abstract class. Then, you can remove the Fly() method from the Boat and Ship classes. This is reasonable too. If you do this, the Boat and Ship classes can use the base class’s Float() method without overriding this method inside them. (Obviously, they can override the behavior if they want.) Consider airplanes now. You cannot remove the Fly() method from the Airplane class. So, you can see that if you need to add a new type of vehicle with a different behavior, the maintenance of the code will become tough. When you have a ship, boat, and airplane, you may find that placing the Float() method in an abstract class is beneficial. But if you have a ship, boat, airplane, helicopter, and rocket, you may find that having the Fly() method in the abstract class instead is more beneficial for you. It may not always be easy to determine which behavior(s) you should consider as common behaviors (particularly in a growing application that keeps adding different vehicles). 23

Chapter 2

Abstract Class or Interface?

It is not the only issue to consider. Later, you’ll see the SOLID principles (Chapter 4) and you will learn that it is not a good idea to put many different behaviors in a single class, even if this design may seem appealing when you have so many common behaviors among many different classes. Now, go back to the initial code segment where you considered only boats and airplanes. In this case, if you use an interface, you need to implement all the interface methods. As a result, since boats do not fly, you needed to override the Fly() method in the Boat class as follows: public void Fly() {     throw new NotImplementedException(); } Again, airplanes do not float in normal situations. So, you need to override the method as follows: public void Float() {     throw new NotImplementedException(); } This kind of code creates a problem when you try to use polymorphic code. These implementations can throw exceptions when you iterate over the vehicles using supertype references and try to access the fly or float behaviors. For example, the following code raises an exception: List vehicles = new List() {   new Boat(),   new Airplane() };   foreach( Vehicle vehicle in vehicles ) {   vehicle.Float();   vehicle.Fly(); }

24

Chapter 2

Abstract Class or Interface?

Note  In Chapter 4, when I discuss the Liskov substitution principle (LSP), you’ll see a detailed discussion of this. Apart from the issues just mentioned, consider some unusual situations such as an airplane that can float. Or, considering technology enhancements, we might expect to see flying cars someday soon. These considerations give you a clue that separating behavior from a vehicle can help you maintain the application. So, instead of writing a complete program that follows the initial design, let us jump to the next section, where we will start with a better approach.

B  etter Program Let us suppose each vehicle should have a registration number from the government of the country. In this case, you’re likely to use this field inside an abstract class. But if you need to consider the different types of vehicles, such as airplanes, ships, or boats, that can show different behaviors, an interface is a better choice. What can you do now? Your guess is correct. You can combine an abstract class with an interface in your application. As discussed before, you separate the vehicle behavior and form a different inheritance hierarchy. This kind of design helps you to add dynamic behavior to a vehicle. Look at the upcoming demonstration.

D  emonstration In this demonstration, you can see two different inheritance hierarchies. Here are the important considerations: •

Each vehicle can have different behavior. All these behaviors form one hierarchy. I assume that initially, a vehicle cannot do anything special. To represent this, I add a DoNothing behavior in the born state. In a later stage, the float or fly capability can be added to a vehicle. To represent these two behaviors, I use the FloatCapability and FlyCapability classes, respectively. So, I have an inheritance hierarchy where you see an interface ICapability with three different classes: FloatCapability, FlyCapability, and DoNothing.

The following class diagram shows this inheritance chain (Figure 2-4). 25

Chapter 2

Abstract Class or Interface?

Figure 2-4.  All the possible vehicle behaviors form one inheritance hierarchy The following code segment represents this inheritance chain: interface ICapability {     void CurrentCapability(); } class FloatCapability : ICapability {     public void CurrentCapability()     {         Console.WriteLine("It can float now.");     } } class FlyCapability : ICapability {     public void CurrentCapability()     {         Console.WriteLine("It can fly now.");     } } 26

Chapter 2

Abstract Class or Interface?

class DoNothing : ICapability {     public void CurrentCapability()     {         Console.WriteLine("It does nothing.");     } } I place the different vehicles in their own separate hierarchies. This time I start with an abstract class called Vehicle. The Boat and Airplane classes derive from this class. Here are the additional considerations: •

I assume each vehicle has a registration number, and a vehicle can show just one behavior at a particular time. But you can change the behavior if you want.



To set a particular behavior (or, capability), I use the SetVehicleBehavior() method. To show the current details of a vehicle, there is a DisplayDetails() method.



I place these methods and fields in the abstract class Vehicle, which is as follows: abstract class Vehicle     {         protected string vehicleType = String.Empty;         protected ICapability vehicleBehavior;         protected string registrationNumber = String.Empty;         public abstract void SetVehicleBehavior(ICapability behavior);         public abstract void DisplayDetails();     }

Note Notice that the SetVehicleBehavior(...) method accepts a polymorphic argument that is nothing but a vehicle behavior. I highlight it in bold.

27

Chapter 2



Abstract Class or Interface?

I mentioned earlier that in the born state, a vehicle does not have any special behavior. So, I add the DoNothing behavior to it in the born state. To illustrate this, here is sample code for you from the Boat class constructor: public Boat(string registrationId) {     this.registrationNumber = registrationId;     this.vehicleType = "Boat";     this.vehicleBehavior = new DoNothing(); }

The following class diagram summarizes the detail (Figure 2-5):

Figure 2-5.  Vehicle, Airplane, and Boat form an inheritance hierarchy

28

Chapter 2

Abstract Class or Interface?

POINT TO REMEMBER I remind you that if by mistake, you place the instance fields vehicleType, vehicleBehavior, and registrationNumber in an interface, you will see the compile-time error (CS 0525) saying: Interfaces cannot contain instance fields. So, if you want to use instance fields, you will need an abstract class. Now, go through the complete implementation and output. using System; namespace VehicleDemo {     interface ICapability     {         void CurrentCapability();     }     class FloatCapability : ICapability     {         public void CurrentCapability()         {             Console.WriteLine("It can float now.");         }     }     class FlyCapability : ICapability     {         public void CurrentCapability()         {             Console.WriteLine("It can fly now.");         }     }

29

Chapter 2

Abstract Class or Interface?

    class DoNothing : ICapability     {         public void CurrentCapability()         {             Console.WriteLine("It does nothing.");         }     }     abstract class Vehicle     {         protected string vehicleType = String.Empty;         protected ICapability vehicleBehavior;         protected string registrationNumber = String.Empty;         public abstract void SetVehicleBehavior(ICapability behavior);         public abstract void DisplayDetails();     }     class Boat:Vehicle     {         public Boat(string registrationId)         {             this.registrationNumber = registrationId;             this.vehicleType = "Boat";             this.vehicleBehavior = new DoNothing();         }         public override void SetVehicleBehavior(ICapability behavior)         {             this.vehicleBehavior = behavior;         }         public override void DisplayDetails()         {             Console.WriteLine("Current status of the boat:");

30

Chapter 2

Abstract Class or Interface?

            Console.WriteLine($"Registration number:{this. registrationNumber}");             vehicleBehavior.CurrentCapability();         }     }     class Airplane : Vehicle     {         public Airplane(string registrationId)         {             this.registrationNumber = registrationId;             this.vehicleType = "Airplane";             this.vehicleBehavior = new DoNothing();         }         public override void SetVehicleBehavior(ICapability behavior)         {             this.vehicleBehavior = behavior;         }         public override void DisplayDetails()         {             Console.WriteLine("Current status of the airplane:");             Console.WriteLine($"Registration number: {this. registrationNumber}");             vehicleBehavior.CurrentCapability();         }     }     class Program     {         static void Main(string[] args)         {             try             {                 Console.WriteLine("***Vehicles demo.***");                 Vehicle vehicle = new Boat("B001");                 vehicle.DisplayDetails();                 Console.WriteLine("****************"); 31

Chapter 2

Abstract Class or Interface?

                ICapability currentCapability = new FloatCapability();                 vehicle.SetVehicleBehavior(currentCapability);                 vehicle.DisplayDetails();                 Console.WriteLine("****************");                 vehicle = new Airplane("A002");                 currentCapability = new FlyCapability();                 vehicle.SetVehicleBehavior(currentCapability);                 vehicle.DisplayDetails();                 Console.WriteLine("****************");                 Console.WriteLine("Adding float behavior to the airplane.");                 // Adding float capability to an airplane                 currentCapability = new FloatCapability();                 vehicle.SetVehicleBehavior(currentCapability);                 vehicle.DisplayDetails();                 Console.WriteLine("****************");                 Console.ReadKey();             }             catch( Exception ex)             {                 Console.WriteLine($"Error:{ex}");             }         }     } }

Output Here is the output: ***Vehicles demo.*** Current status of the boat: Registration number: B001 It does nothing.

32

Chapter 2

Abstract Class or Interface?

**************** Current status of the boat: Registration number: B001 It can float now. **************** Current status of the airplane: Registration number: A002 It can fly now. **************** Adding float behavior to the airplane. Current status of the airplane: Registration number: A002 It can float now. ****************

Analysis The competitor to inheritance is composition. When you use object composition, you do a HAS-A test. For example, a car HAS-A body, or a human body HAS-A head. In programming, suppose you represent the human body with a class called HUMAN and a human head with a class HEAD. To represent the line: “A human body has a head,” you would create a HEAD reference inside the HUMAN class. You have noticed that in our code example, each vehicle has different behaviors too. How did we represent these behaviors? There is a separate inheritance chain for each vehicle, and all these behaviors implemented the ICapability interface. Notice that the Vehicle class contains an ICapability reference. This helps a vehicle to show the correct behavior in a particular moment. You also saw that each vehicle can change its behavior at run-time. To achieve these functionalities, you ensured that each behavior implemented the behavior interface properly. This example shows you that by combining the real power of an abstract class, interfaces, and object composition you can make an efficient application.

33

Chapter 2

Abstract Class or Interface?

Summary If you want to have a centralized behavior, use an abstract class. But when you want class-specific implementations, use an interface. This chapter answered the following questions:

34



When is an abstract class a better choice than an interface?



When is an interface a better choice than an abstract class?



What is a diamond problem?



How can you do a HAS-A test?



How can object composition offer a better solution?



How can you change an object’s behavior at runtime?



How can you combine an abstract class with an interface to make an efficient application?

CHAPTER 3

Wise Use of Code Comments Comments help you understand other people’s code. They can describe the program logic. Expert programmers are particular about comments, however. They do not like to see unnecessary comments, for various reasons, which we will discuss. You may or may not agree with all these points of view. The different case studies included in this chapter can help you decide whether to put a comment in your application.

R  ecap It is standard practice to use comments in your program. The C# compiler ignores the comments, but they can help others to understand your code better. Let us consider a real-life scenario. In a software organization, a group of people creates software for its customers. It is possible that after some years, none of them are available. Either these members have moved into a different team, or they have left the organization. In such a case, someone needs to maintain the software and continue fixing the bugs for the customers. But it is very difficult to understand the code if there is no hint or explanation about the program logic. Comments are useful in such scenarios. In C#, you see the following type of comments: Type-1: Single-line comments using double forward-slash (//). Here is a code segment that starts with a single-line comment. // Testing whether 2 is greater than 1 Console.WriteLine(2 > 1); Type-2: You can use multi-line comments to comment on more than one line at a time. You use this to comment out a block of statements. Here is a code segment that starts with multi-line comments. © Vaskaran Sarcar 2021 V. Sarcar, Simple and Efficient Programming with C#, https://doi.org/10.1007/978-1-4842-7322-7_3

35

Chapter 3

Wise Use of Code Comments

/* Now I use multi-line comments. It spans multiple lines. Here I multiply 2 with 3. */ Console.WriteLine(2 * 3); Type-3: These are documentation comments, which are special comments that contain XML text. There are two types: They either start with three slashes(///), often called single-line documentation comments, or they are delimited comments that start with a slash and two stars(/**). Here is a code segment that uses single-line documentation comments. /// /// This is a custom class. ///
There is no method inside it.
/// class MyClass { } Here is a code segment that uses a different form: /** * * This is another custom class. *
It is also empty now.
* */ class MyAnotherClass { } In the end, the aim is the same: comments help others to understand why you wrote a piece of code.

36

Chapter 3

Wise Use of Code Comments

IN A NUTSHELL • Comments are simple notes or some text. You use them for human readers, not for the C# compiler. The C# compiler ignores the text inside a comment block. • In the software industry, many technical reviewers review your code. The comments help them understand the program logic. • A developer can forget the logic after some months too. These comments can help him remember his logic.

Initial Program You know that when the C# compiler sees a comment, it ignores it. Demonstration 1 is a complete program with many different comments. Compile and run this program to ensure you see the expected output.

Demonstration 1 In this program, you calculate the area of a rectangle. using System; namespace Demonstration1 {     ///     /// This is the Rectangle class     ///     class Rectangle     {         readonly double l; // length of the rectangle         readonly double b; // breadth of the rectangle         public Rectangle(double le, double br)         {             l = le;             b = br;         } 37

Chapter 3

Wise Use of Code Comments

        // Measuring the area         public double Area()         {             return l * b;         }     }     class Program     {         static void Main(string[] args)         {             Console.WriteLine("***Measuring the area of a rectangle.***");             Rectangle r = new Rectangle(2.5, 10);             double area = r.Area();             Console.WriteLine($"The area of the rectangle is {area} square units.");             Console.ReadKey();         }     } }

Output Here is the output: ***Measuring the area of a rectangle.*** The area of the rectangle is 25 square units.

Analysis This program uses different types of comments to explain the code. These are not doing any harm to the program. Now the question is: are they necessary? You will find many people in the software industry who dislike comments. They believe that common people do not read your code. In general, a programmer or a developer reads your code. So, using too much comment in a program is unnecessary. Also, older comments can be

38

Chapter 3

Wise Use of Code Comments

misleading if you do not maintain them. My personal belief is that comments are good if they are necessary. I dislike unnecessary comments and like to remove them if my code is expressive enough.

Better Program Can you rewrite the program in Demonstration 1 without comments? Yes, you can. You can delete all the comments. Then you can compile and run the program to confirm you get the same output. But the question is: When you do that, will your code be readable? Can a person understand it easily? Let’s have a look at Demonstration 2.

Demonstration 2 This is a modified version of Demonstration 1. What are the changes? Notice that I have renamed the variables and the area method inside the Rectangle class. These new names are expressive enough. Anyone who reads this code should have a good idea about what I aim to do. using System; namespace Demo1Modified {     class Rectangle     {         readonly double length;         readonly double breadth;         public Rectangle(double length, double breadth)         {             this.length = length;             this.breadth = breadth;         }         public double RectangleArea()         {             return length * breadth;         }     } 39

Chapter 3

Wise Use of Code Comments

    class Program     {         static void Main(string[] args)         {             Console.WriteLine("***Measuring the area of a rectangle.***");             Rectangle rectangleObject = new Rectangle(2.5, 10);             double area = rectangleObject.RectangleArea();             Console.WriteLine($"The area of the rectangle is {area} square units.");             Console.ReadKey();         }     } }

Analysis This demonstration is easy to understand. Have you noticed that this time I chose variable names length and breadth? In Demonstration 1, I used l (lowercase L, not 1) and b, respectively. And to let others understand this code, I needed to write inline comments such as // length of the rectangle or // breadth of the rectangle. Similarly, when I choose the method name RectangleArea(), one can presume what this method is about to do. A similar type of method name would be useful if you were to work on areas of different shapes, such as circles, squares, or triangles.

Use the Power of C# Sometimes you see comments that appear very helpful at the beginning. Consider the following class that consists of a TODO comment saying you do not intend to use the SayHello() method in the future. It also suggests using SayHi() starting with the next release. class SimpleTodo     {     // TODO-We'll replace this method shortly.     // Use SayHi() from the next release(Version-2.0). 40

Chapter 3

Wise Use of Code Comments

        public void SayHello()         {             Console.WriteLine("Hello, Reader!");         }         public void SayHi()         {             Console.WriteLine("Hi, Reader!");             Console.WriteLine("This is the latest method.");         }     } This TODO comment seems to be useful. Now see a sample client code that uses these methods: class Program { static void Main(string[] args)   {    Console.WriteLine("***TODO comments example.***");    SimpleTodo simpleTodo = new SimpleTodo();    simpleTodo.SayHello();    simpleTodo.SayHi();    Console.ReadKey();   } } This client code is simple. There is nothing special in this code. Think from a company perspective now: a company does not share the actual code with its clients. Instead, the company tells the client how to use the functionalities of the application. But how will the client know that you do not intend to use SayHello() from version 2.0 onward? One way is to include this information in the user manual. But there is an alternative approach too. You can use the power of attributes. It is better in the sense that human behavior often resists change. If they can do the work using an old method, it is very likely that they will be too lazy to test the new method. Here is a sample:

41

Chapter 3

Wise Use of Code Comments

class SimpleTodo { [ObsoleteAttribute("This method is obsolete.Call SayHi() instead.")] public void SayHello() {   Console.WriteLine("Hello, Reader!"); } public void SayHi() {   Console.WriteLine("Hi, Reader!");   Console.WriteLine("This is the latest method."); } } Now the same client code can tell that SayHello() is obsolete and a client should use SayHi() instead of this old method. Figure 3-1 is a screenshot from the Visual Studio IDE.

Figure 3-1.  The SayHello() method is obsolete

42

Chapter 3

Wise Use of Code Comments

Summary The famous book Clean Code (Pearson Education, Inc.) by Robert C. Martin tells us, “Comments are always failures. We must have them because we cannot always figure out how to express ourselves without them, but their use is not a cause for celebration.” This book continues: “Every time you express yourself in code, you should pat yourself on the back. Every time you write a comment, you should grimace and feel the failure of your ability of expression.” Another great book, The Pragmatic Programmer by Andrew Hunt and David Thomas, tells us: “Programmers are taught to comment their code: good code has lots of comments. Unfortunately, they are never taught why code needs comments: bad code requires lots of comments.” You may or may not always agree with these thoughts, and you will find people who can point to pros and cons on both sides. Even these books show some nice examples of both good and bad comments. There are plenty of examples where the actual code is tricky or difficult to understand. Some well-maintained good comments can help a first-time reader/ developer in those contexts. For me, it is helpful when I hover my mouse on an in-built function to understand it better. For example, in this book in many examples, I have generated some random numbers. There are overloaded methods to do this activity. I often use the following form. The associated comments are straightforward for me to understand how this method works. I pick the following in-built comments that are associated with a particular version of the Next method. // // Summary: //     Returns a non-negative random integer that is less than the //     specified maximum. // // Parameters: //   maxValue: //     The exclusive upper bound of the random number to be generated. //     maxValue must //     be greater than or equal to 0. //

43

Chapter 3

Wise Use of Code Comments

// Returns: //     A 32-bit signed integer that is greater than or equal to 0, and less //     than maxValue; //     that is, the range of return values ordinarily includes 0 but not //     maxValue. However, //     if maxValue equals 0, maxValue is returned. // // Exceptions: //   T:System.ArgumentOutOfRangeException: //     maxValue is less than 0. public virtual int Next(int maxValue); This is the reason I suggest you take a close look and analyze more before you place a comment in your code. Use them when they are truly beneficial. You will get a better idea of what is useful when your code goes through peer reviews.

Note  You should be careful. A line of commented code becomes worse when you see a method or variables that are no longer in use. It can also cause trouble when you use a comment that does not stay close to the actual code. In the worst case, you may see comments that give you incorrect information. At any cost, you should not allow bad or unnecessary comments to live in your application. One final point: comments are not always used to describe the code. You may see commented-out code in an application too. Keeping commented-out code is not a recommended practice. But I use some commented code in my books for further demonstration purposes. For example, I may point out an alternative way of calling a method. Sometimes I keep the commented code with possible output to show you why this is correct or incorrect. But following the experts’ suggestion, I do not like to see commented code in an enterprise application. It is because if needed, you can always find the old code via a source code version management tool such as Git or SVN.

44

Chapter 3

Wise Use of Code Comments

In short, this chapter discussed the following questions: •

What are code comments?



What are the different types of comments?



Why are good comments beneficial?



Why are unnecessary comments bad and how can you avoid them?



How can you avoid plain comments using C# attributes?

45

PART II

Important Principles Part II consists of two chapters, in which we’ll examine the use of the following: •

SOLID principles. These are a combination of five design guidelines.



Don’t Repeat Yourself (DRY) principle

In an object-oriented programming world, there is no shortage of principles, but those we will discuss in the next two chapters are the fundamental design guidelines for a better application. One cannot predict all future requirements in advance, so changes are often needed in enterprise applications. This is why a flexible application that can adopt future requirements easily is considered a better application. This part will review case studies using (and without using) these principles and help you think about their importance. A detailed study of these principles can help you create efficient and flexible applications.

CHAPTER 4

Know SOLID Principles C# is a powerful language. It supports object-oriented programming and has myriad features. If we compare it with the old days, it seems that coding has become easy with the support of these powerful features. But the hard truth is: simply using these features in an application does not guarantee that you have used them in the right way. For any given requirement, it is vital to identify classes, objects, and how they communicate with each other. In addition, your application must be flexible and extendable to allow for future enhancements. Now the question is: what are the specific guidelines? You need to follow experts’ footprints. Robert Cecil Martin is a famous name in the programming world. He is an American software engineer and best-selling author, and is also known as “Uncle Bob.” He promoted many principles, a subset of which are the following: •

Single Responsibility Principle (SRP)



Open/Closed Principle (OCP)



Liskov Substitution Principle (LSP)



Interface Segregation Principle (ISP)



Dependency Inversion Principle (DIP)

Robert C. Martin and Micah Martin discussed these principles in their book Agile Principles, Patterns and Practices in C# (Prentice Hall). By taking the first letter of each principle, Michael Feathers introduced the SOLID acronym to aid in remembering these names. Design principles are high-level guidelines that you can use to make better software. These are not bound to any particular computer language. So, if you understand these concepts using C#, you can use them with similar languages like Java or C++. See https://sites.google.com/site/unclebobconsultingllc/getting-­a-­solid-­start to learn the thoughts of Robert C. Martin about this:

© Vaskaran Sarcar 2021 V. Sarcar, Simple and Efficient Programming with C#, https://doi.org/10.1007/978-1-4842-7322-7_4

49

Chapter 4

Know SOLID Principles

The SOLID principles are not rules. They are not laws. They are not perfect truths. They are statements on the order of “An apple a day keeps the doctor away.” This is a good principle, it is good advice, but it’s not a pure truth, nor is it a rule. —Uncle Bob In this chapter, we’ll explore these principles in detail. In each case, I start with a program that compiles and runs successfully, but that does not follow any specific design guideline. In the analysis section, we’ll discuss the possible drawbacks and try to find a better solution using these principles. This process can help you understand the importance of these design guidelines. I remind you—the purpose of these case studies is to help you think better and create better applications, but these are not rules that you need to follow in every context.

Single Responsibility Principle (SRP) A class acts like a container that can hold many things, such as data, properties, or methods. If you put in too much data, properties, or methods that are not related to each other, you end up with a bulky class that can create problems in the future. Let us consider an example. Suppose you create a class with multiple methods that do different things. In such a case, even if you make just a small change in one method, you need to retest the whole class again to ensure the workflow is correct. Changes in one method can impact the other method(s) in the class. The single responsibility principle (SRP) opposes this idea of putting multiple responsibilities in a class. It says that a class should have only one reason to change. So, before you make a class, identify the responsibility or the purpose of the class. If multiple members help you to achieve one single purpose, you can place all those members inside the class.

POINT TO REMEMBER When you follow SRP, your code is smaller, cleaner, and less fragile. Now the question is: How do you follow this principle? A simple answer is: You can divide a big problem into smaller chunks based on different responsibilities and put each of these small parts into a separate class. The next question is: What do we mean by responsibility? In simple words: responsibility is a reason for a change. 50

Chapter 4

Know SOLID Principles

In the upcoming discussion, you will see a class that contains three different methods that are not closely related to each other. In the end, I segregate the code based on different responsibilities and put them into different classes. Let’s start.

Initial Program In Demonstration 1, there is an Employee class with three different methods. Here are the details: •

The DisplayEmployeeDetail() shows the employee’s name and his working experience in years.



The CheckSeniority() method can evaluate whether an employee is a senior person. I assume that if the employee has 5+ years of experience, he is a senior employee; otherwise, he is a junior employee.



The GenerateEmployeeId() method generates an employee ID using string concatenation. The logic is simple: I concatenate the first word of the first name with a random number to form an employee ID. In the following demonstration, inside Main(), I create two Employee instances and use these methods to display the relevant details.

Demonstration 1 Here is a program that does not follow SRP. using System; namespace WithoutSRPDemo {     class Employee     {         public string empFirstName, empLastName, empId;         public double experienceInYears;          public Employee(string firstName, string lastName, double experience)         {             this.empFirstName = firstName;             this.empLastName = lastName; 51

Chapter 4

Know SOLID Principles

            this.experienceInYears = experience;         }         public void DisplayEmployeeDetail()         {             Console.WriteLine($"The employee name: {empLastName}, {empFirstName}");             Console.WriteLine($"This employee has {experienceInYears} years of experience.");         }         public string CheckSeniority(double experienceInYears)         {             if (experienceInYears > 5)                 return "senior";             else                 return "junior";         }         public string GenerateEmployeeId(string empFirstName)         {             int random = new System.Random().Next(1000);             empId = String.Concat(empFirstName[0], random);             return empId;         }     }     class Program     {         static void Main(string[] args)         {             Console.WriteLine("*** A demo without SRP.***");             Employee robin = new Employee("Robin", "Smith", 7.5);             robin.DisplayEmployeeDetail();             string empId = robin.GenerateEmployeeId(robin.empFirstName);             Console.WriteLine($"The employee id: {empId}");

52

Chapter 4

Know SOLID Principles

            Console.WriteLine($"This employee is a " +                 $"{robin.CheckSeniority(robin.experienceInYears)} employee.");             Console.WriteLine("\n*******\n");             Employee kevin = new Employee("Kevin", "Proctor", 3.2);             kevin.DisplayEmployeeDetail();             empId = kevin.GenerateEmployeeId(kevin.empFirstName);             Console.WriteLine($"The employee id: {empId}");             Console.WriteLine($"This employee is a " +                 $"{kevin.CheckSeniority(kevin.experienceInYears)} employee.");             Console.ReadKey();         }     } }

Output Here is a sample output (the employee ID can vary in your case). *** A demo without SRP.*** The employee name: Smith, Robin This employee has 7.5 years of experience. The employee id: R586 This employee is a senior employee. ******* The employee name: Proctor, Kevin This employee has 3.2 years of experience. The employee id: K459 This employee is a junior employee.

53

Chapter 4

Know SOLID Principles

Analysis What is the problem with this design? This answer is that I violate the SRP here. Displaying an employee detail, generating an employee ID, and checking seniority level are all different activities. Since I put everything into a single class, I may face problems when adopting changes in the future, such as if top management sets a different criterion to decide a seniority level. One could also use a complex algorithm to generate the employee ID. In each case, you’d need to modify the Employee class and so forth. You can now see that it is better to follow the SRP and separate the activities.

Better Program In the following demonstration, I introduce two more classes. The SeniorityChecker class now contains the CheckSeniority() method, and the EmployeeIdGenerator class contains the GenerateEmployeeId() method to generate the employee ID. As a result, in the future, if you need to change the program logic to determine the seniority level, or use a new algorithm to generate an employee ID, you can make the changes in the respective classes. Other classes are untouched, so you do not need to retest them. This time, I improve code readability too. Notice that in Demonstration 1, I called all the required methods inside Main(). But for better readability and to avoid clumsiness inside Main(), this time I introduce three static methods: PrintEmployeeDetail(...), PrintEmployeeId(...), and PrintSeniorityLevel(...). These methods call the DisplayEmployeeDetail() method, GenerateEmployeeId() method, and CheckSeniority() method, respectively. These three methods are not necessary, but they make the client code simple and easily understandable.

Demonstration 2 Here is the complete demonstration that follows SRP: using System; namespace SRPDemo {     class Employee     {         public string empFirstName, empLastName; 54

Chapter 4

Know SOLID Principles

        public double experienceInYears;          public Employee(string firstName, string lastName, double experience)         {             this.empFirstName = firstName;             this.empLastName = lastName;             this.experienceInYears = experience;         }         public void DisplayEmployeeDetail()         {             Console.WriteLine($"The employee name: {empLastName}, {empFirstName}");             Console.WriteLine($"This employee has {experienceInYears} years of experience.");         }     }     class SeniorityChecker     {         public string CheckSeniority(double experienceInYears)         {             if (experienceInYears > 5)                 return "senior";             else                 return "junior";         }     }     class EmployeeIdGenerator     {         string empId;         public string GenerateEmployeeId(string empFirstName)         {             int random = new System.Random().Next(1000);             empId = String.Concat(empFirstName[0], random);             return empId;         }     } 55

Chapter 4

Know SOLID Principles

    class Program     {         static void Main(string[] args)         {             Console.WriteLine("*** A demo that follows SRP.***");             Employee robin = new Employee("Robin", "Smith", 7.5);             PrintEmployeeDetail(robin);             PrintEmployeeId(robin);             PrintSeniorityLevel(robin);             Console.WriteLine("\n*******\n");             Employee kevin = new Employee("Kevin", "Proctor", 3.2);             PrintEmployeeDetail(kevin);             PrintEmployeeId(kevin);             PrintSeniorityLevel(kevin);             Console.ReadKey();         }         private static void PrintEmployeeDetail(Employee emp)         {             emp.DisplayEmployeeDetail();         }         private static void PrintEmployeeId(Employee emp)         {             EmployeeIdGenerator idGenerator = new EmployeeIdGenerator();              string empId = idGenerator.GenerateEmployeeId(emp.empFirstName);             Console.WriteLine($"The employee id: {empId}");         }         private static void PrintSeniorityLevel(Employee emp)         {             SeniorityChecker seniorityChecker = new SeniorityChecker();             string seniorityLevel = seniorityChecker.CheckSeniority( emp.experienceInYears); 56

Chapter 4

Know SOLID Principles

            Console.WriteLine($"This employee is a {seniorityLevel} employee.");         }     } }

Output Here is the output. Notice that it is similar to the previous output, except that the first line states that this program follows the SRP now. (The employee ID can vary in your case). *** A demo that follows SRP.*** The employee name: Smith, Robin This employee has 7.5 years of experience. The employee id: R841 This employee is a senior employee. ******* The employee name: Proctor, Kevin This employee has 3.2 years of experience. The employee id: K676 This employee is a junior employee.

POINT TO NOTE Note that the SRP does not insist that a class should have at most one method. Here the emphasis is on the single responsibility. There may be closely related methods that can help you to implement a responsibility. For example, if you have different methods to display the first name, the last name, and a full name, you can put these methods in the same class. These methods are closely related, and it makes sense to place all these display methods inside the same class.

57

Chapter 4

Know SOLID Principles

In addition, you should not conclude that you must always separate responsibilities in every application that you make. You need to analyze the change’s nature. Having too many classes can make your application complex, which is difficult to maintain. But if you know this principle and think carefully before you implement a design, you are likely to avoid the mistakes that I discussed earlier.

Open/Closed Principle (OCP) In this section, we’ll examine the open/closed principle (OCP) in detail. It originated from the work of Bertrand Meyer.

Of all the principles of object-oriented design, this is the most important. —Robert C. Martin This principle says that a software entity (a class, module, method, etc.) should be open for extension, but closed for modification. The idea behind this design philosophy is that in a stable and working application, once you create a class and other parts of your application start using it, any further change in the class can cause the working application to break. If you require new features (or functionalities), instead of changing the existing class, you can extend it to adopt those new requirements. What is the benefit? Since you do not change the old code, your old functionalities continue to work without any problem, and you can avoid testing them again. Instead, you need to test the “extended” part (or functionalities) only. In 1988, Bertrand Meyer suggested the use of inheritance in this context, saying the following: “A class is closed, since it may be compiled, stored in a library, baselined, and used by client classes. But it is also open, since any new class may use it as a parent, adding new features. When a descendant class is defined, there is no need to change the original or to disturb its clients.” But inheritance promotes tight coupling. In programming, we like to remove these tight couplings. Robert C. Martin improved the definition and made it polymorphic OCP. The new proposal uses abstract base classes, which use protocols instead of a superclass to allow different implementations. These protocols are closed for modification and provide another level of abstraction, which enables loose coupling. In this chapter, we follow Martin’s idea that promotes polymorphic OCP. 58

Chapter 4

Know SOLID Principles

Note  In the final chapter of this book, I described some common terms including “cohesion” and “coupling.” If required, you can have a quick look at them now.

Initial Program Assume that there is a small group of students who appear in a certification examination. To demonstrate this, I choose a small number of participants. The small size helps you to focus on the principle, not unnecessary details. Sam, Bob, John, and Kate are the four students in this example. They all belong to the Student class. To make a Student class instance, you supply a name, registration number, and grades obtained in the examination. You also mention whether a student belongs to the science department or the arts department. So, you will see the following lines of code in the upcoming example: Student Student Student Student

sam = new Student("Sam", "R001", 81.5,"Science"); bob = new Student("Bob", "R002", 72,"Science"); john = new Student("John", "R003",71,"Arts"); kate = new Student("Kate", "R004", 66.5,"Arts");

Suppose you start with two instance methods. DisplayResult() displays the result with all necessary details of a student, and the EvaluateDistinction() method evaluates whether a student is eligible for a distinction certificate. I assume that if a student from the science department scores above 80 in this examination, he or she gets the certificate with distinction. But the criterion for a student from the arts department is slightly relaxed. An arts student gets the distinction if his or her score is above 70. I assume that you understand SRP at this point, so you know you should not place DisplayResult() and EvaluateDistinction() in the same class, like the following:     class Student     {         readonly string         readonly string         readonly string         readonly double

name; registrationNumber; department; score;

59

Chapter 4

Know SOLID Principles

        public Student(string name, string registrationNumber, double score, string department)         {             this.name = name;             this.registrationNumber = registrationNumber;             this.score = score;             this.department = department;         }         public void DisplayResult()         {             Console.WriteLine($"Name:{this.name} \nReg Number:{ this.registrationNumber} " +                 $"\nDept:{this.department} \nscore: {this.score}");             Console.WriteLine("*************");         }         public void EvaluateDistinction()         {             if (this.score > 80 && this.department == "Science")             {                 Console.WriteLine($"{this.registrationNumber} has passed with distinction.");             }             if (this.score > 70 && this.department == "Arts")             {                 Console.WriteLine($"{this.registrationNumber} has passed with distinction.");             }         }     } What are the problems in this code segment? •

60

First, notice that I violate the SRP when I place both the DisplayResult() and the EvaluateDistinction() methods inside the Student class.

Chapter 4

Know SOLID Principles



In the future, the examining authority can change the distinction criteria. In this case, you need to change the EvaluateDistinction() method. Does this code solve the problem? In the current situation, the answer is yes. But a college authority could change the distinction criteria again. How many times will you modify the EvaluateDistinction() method?



Remember that each time you modify the method, you need to write/ modify the existing test cases too.

You can see that every time the distinction criteria changes, you need to modify the EvaluateDistinction() method in the Student class. So, this class does not follow SRP and is also not closed for modification. Once you understand these problems, you can start a better design that follows SRP. Here are the main characteristics of the design: •

In the following program, Student and DistinctionDecider are two different classes.



The DistinctionDecider class contains the EvaluateDistinction() method.



To show details of a student, you can override the ToString() method, instead of using a separate method DisplayResult(). So, inside the Student class, you see the ToString() method now.



Inside Main(), you see the following line: List enrolledStudents = MakeStudentList(); The MakeStudentList() method creates a list of students. It helps to avoid repetitive code for each student. You pass this list inside DisplayStudentResults() to print the students’ details one by one. You also use the same list to invoke EvaluateDistinction() to identify the students who have gotten the distinction.

Demonstration 3 Here is the complete demonstration.

61

Chapter 4

Know SOLID Principles

using System; using System.Collections.Generic; namespace WithoutOCPDemo {     class Student     {         internal string name;         internal string registrationNumber;         internal string department;         internal double score;         public Student(string name,                        string registrationNumber,                        double score,                        string department)         {             this.name = name;             this.registrationNumber = registrationNumber;             this.score = score;             this.department = department;         }         public override string ToString()         {             return ($"Name: {this.name} " +                 $"\nReg Number: {this.registrationNumber} " +                 $"\nDept: {this.department} " +                 $"\nscore: {this.score}" +                 $"\n*******");         }     }     class DistinctionDecider     {         public void EvaluateDistinction(Student student)         {

62

Chapter 4

Know SOLID Principles

            if (student.department == "Science")             {                 if (student.score > 80)                 {                     Console.WriteLine($"{student.registrationNumber} has received a distinction in science.");                 }             }             if (student.department == "Arts")             {                 if (student.score > 70)                 {                     Console.WriteLine($"{student.registrationNumber} has received a distinction in arts.");                 }             }         }     }     class Program     {         static void Main(string[] args)         {             Console.WriteLine("*** A demo without OCP.***");             List enrolledStudents = MakeStudentList();             // Display results.             Console.WriteLine("===Results:===");             foreach(Student student in enrolledStudents)             {                 Console.WriteLine(student);             }             // Evaluate distinctions.              DistinctionDecider distinctionDecider = new DistinctionDecider();             Console.WriteLine("===Distinctions:==="); 63

Chapter 4

Know SOLID Principles

            foreach (Student student in enrolledStudents)             {                 distinctionDecider.EvaluateDistinction(student);             }             Console.ReadKey();         }         private static List MakeStudentList()         {             Student sam = new Student("Sam", "R001", 81.5, "Science");             Student bob = new Student("Bob", "R002", 72, "Science");             Student john = new Student("John", "R003", 71, "Arts");             Student kate = new Student("Kate", "R004", 66.5, "Arts");             List students = new List();             students.Add(sam);             students.Add(bob);             students.Add(john);             students.Add(kate);             return students;         }     } }

Output Here is the output: *** A demo without OCP.*** ===Results:=== Name: Sam Reg Number: R001 Dept: Science score: 81.5 *******

64

Chapter 4

Know SOLID Principles

Name: Bob Reg Number: R002 Dept: Science score: 72 ******* Name: John Reg Number: R003 Dept: Arts score: 71 ******* Name: Kate Reg Number: R004 Dept: Arts score: 66.5 ******* ===Distinctions:=== R001 has received a distinction in science. R003 has received a distinction in arts.

Analysis Now you have followed the SRP. If, in the future, the examining authority changes the distinction criteria, you do not have to touch the Student class. So, this part is closed for modification. It solves one part of the problem. Now think about another future possibility: the college authority can introduce a new stream, such as commerce, and set a new distinction criterion for this stream. You need to make some obvious changes again. For example, you need to modify the EvaluateDistinction() method and add another if statement to consider commerce students. Now the question is: Is it acceptable to modify the EvaluateDistinction() method in this manner? Remember that each time you modify the method, you need to write/modify the entire code workflow again. You understand the problem now. In this demonstration, every time the distinction criteria changes, you need to modify the EvaluateDistinction() method in the DistinctionDecider class. So, this class is not closed for modification.

65

Chapter 4

Know SOLID Principles

Better Program To tackle this problem, you can write a better program. The following program shows such an example and follows the OCP principle that suggests we write code segments (such as classes or methods) that are open for extension but closed for modification.

Note OCP can be achieved in different ways, but abstraction is the heart of this principle. If you can design your application following OCP, your application will be flexible and extensible. It is not always easy to fully implement this principle, but partial OCP compliance too can generate greater benefit to you. Also notice that I started Demonstration 3 by following SRP. If you do not follow OCP, you may end up with a class that performs multiple tasks, which means the SRP is broken. This time, we need to tackle the evaluation method for distinction in a better way. So, I create an interface IDistinctionDecider that contains a method EvaluateDistinction. Here is the interface: interface IDistinctionDecider {     void EvaluateDistinction(Student student); } The ArtsDistinctionDecider and ScienceDistinctionDecider implement this interface and override the IDistinctionDecider method to serve their purpose. Here is the code segment for this. The different criteria for each class are shown in bold: class ArtsDistinctionDecider : IDistinctionDecider { public void EvaluateDistinction(Student student)   {    if (student.score > 70)    {     Console.WriteLine($"{student.registrationNumber} has got distinction in arts.");    }   } } 66

Chapter 4

Know SOLID Principles

class ScienceDistinctionDecider : IDistinctionDecider { public void EvaluateDistinction(Student student) {   if (student.score > 80)   {    Console.WriteLine($"{student.registrationNumber} has distinction in science.");    }   } } The previous code segment clearly shows the distinction criteria in different streams. So, I remove the department field from the Student class now. The remaining code is easy, and you should not have any trouble understanding the following demonstration.

Demonstration 4 Here is the modified program: using System; using System.Collections.Generic; namespace OCPDemo {     class Student     {         internal string name;         internal string registrationNumber;         internal double score;         public Student(string name,                        string registrationNumber,                        double score                        )         {             this.name = name;             this.registrationNumber = registrationNumber; 67

Chapter 4

Know SOLID Principles

            this.score = score;         }         public override string ToString()         {             return(                   $"Name: {this.name} " +                   $"\nReg Number: {this.registrationNumber} " +                   $"\nscore: {this.score}\n*******");             }     }     interface IDistinctionDecider     {         void EvaluateDistinction(Student student);     }     class ArtsDistinctionDecider : IDistinctionDecider     {         public void EvaluateDistinction(Student student)         {             if (student.score > 70)             {                 Console.WriteLine($"{student.registrationNumber} has received a distinction in arts.");             }         }     }     class ScienceDistinctionDecider : IDistinctionDecider     {         public void EvaluateDistinction(Student student)         {             if (student.score > 80)             {                 Console.WriteLine($"{student.registrationNumber} has received a distinction in science.");             }         }     } 68

Chapter 4

Know SOLID Principles

    class Program     {         static void Main(string[] args)         {             Console.WriteLine("*** A demo that follows OCP.***");             List scienceStudents = MakeScienceStudentList();             List artsStudents = MakeArtsStudentList();             // Display results.             Console.WriteLine("===Results:===");             foreach (Student student in scienceStudents)             {                 Console.WriteLine(student);             }             foreach (Student student in artsStudents)             {                 Console.WriteLine(student);             }             // Evaluate distinctions.             Console.WriteLine("===Distinctions:===");             // For science students.             IDistinctionDecider distinctionDecider = new ScienceDistinctionDecider();             foreach (Student student in scienceStudents)             {                 distinctionDecider.EvaluateDistinction(student);             }             // For arts students.             distinctionDecider = new ArtsDistinctionDecider();             foreach (Student student in artsStudents)             {                 distinctionDecider.EvaluateDistinction(student);             }

69

Chapter 4

Know SOLID Principles

            Console.ReadKey();         } private static List MakeScienceStudentList()     {        Student sam = new Student("Sam", "R001", 81.5);        Student bob = new Student("Bob", "R002", 72);        List students = new List();        students.Add(sam);        students.Add(bob);        return students;     }     private static List MakeArtsStudentList()     {      Student john = new Student("John", "R003", 71);      Student kate = new Student("Kate", "R004", 66.5);      List students = new List();      students.Add(john);      students.Add(kate);      return students;     }   } }

Output Notice that the output is the same, except the first line, which states that this program follows the OCP. ***A demo that follows OCP.*** ===Results:=== Name: Sam Reg Number: R001 score: 81.5 *******

70

Chapter 4

Know SOLID Principles

Name: Bob Reg Number: R002 score: 72 ******* Name: John Reg Number: R003 score: 71 ******* Name: Kate Reg Number: R004 score: 66.5 ******* ===Distinctions:=== R001 has received a distinction in science. R003 has received a distinction in arts.

Analysis What are the key advantages now? •

The Student class and IDistinctionDecider both are unchangeable for any future changes in the distinction criteria. They are closed for modification.



Notice that every participant follows the SRP.



If you consider students from a different streams, such as commerce, you can add a new derived class—say, CommerceDistinctionDecider—that implements the IDistinctionDecider interface, and you can set new distinction criteria for commerce students.



Using this approach, you avoid an if-else chain (shown in Demonstration 3). This chain could grow if you consider new streams such as commerce. In cases like this, avoiding a big if-else chain is considered a better practice.

71

Chapter 4

Know SOLID Principles

Note  To describe this principle, I used the concept of class. It should be noted that Robert C. Martin described this principle in terms of modules. The term “module” can be confusing if you purely think in terms of C#. For example, to describe the Module class in System.Reflection, Microsoft documentation says (See https://docs.microsoft.com/en-­us/dotnet/api/system. reflection.module?view=net-­5.0): a module is a portable executable file, such as type.dll or application.exe, consisting of one or more classes and interfaces. This doc also says that a .NET Framework module is not the same as a module in Visual Basic, which is used by programmers to organize functions and subroutines in an application. Similarly, any Python programmer knows that a module can contain many things. For example, to organize his code, he can place variables, functions, and classes in a module. He creates a separate file with a .py extension for this purpose. Later, he can import the whole module or a particular function from the module in his current file. It is time to study the next principle.

Liskov Substitution Principle (LSP) This principle was initially introduced in the work of Barbara Liskov in 1988. The Liskov substitution principle (LSP) says that you should be able to substitute a parent (or base) type with a subtype. In other words, in a program segment, you can use a derived class instead of its base class without altering the correctness of the program. Recall how you use inheritance? There is a base class and you create one (or more) derived classes from it. Then you can add new methods in the derived classes. As long as you directly use the derived class method with a derived class object, everything is fine. A problem may occur though if you try to get polymorphic behavior without following the LSP. How? Assume that there are two classes such that B is the base class and D is the subclass (of B). Furthermore, assume that there is a method that accepts a reference of B as an argument. This method works fine. But the method can misbehave if you pass a D reference instead of B to the method. And this can happen if you do not follow LSP.

72

Chapter 4

Know SOLID Principles

Note  Polymorphic code shows off your expertise, but remember that it’s the developer’s responsibility to implement polymorphic behavior properly and avoid unwanted outcomes. In this chapter, I help you understand this principle with two case studies. In the first case study, I start with a class Rectangle that has a constructor and a method called ShowArea(). Inside the constructor, I display the length and breadth of a Rectangle instance. The ShowArea() method displays the area of a rectangle object. Inside the Main() method, I create two Rectangle instances and invoke the ShowArea()method. In this program, you do not need to supply the breadth for a rectangle. It is because by default it takes two units. So, you can see that both the following lines in this code segment work fine: IRectangle shape = new Rectangle(10, 5.5); shape = new Rectangle(25); Here is complete program: using System; namespace UnderstandingLSP {    class Rectangle     {         protected double length, breadth;         public Rectangle(double length,                         double breadth=2)         {             this.length = length;             this.breadth = breadth;             Console.WriteLine($"Length = {length} units.");             Console.WriteLine($"Breadth = {breadth} units.");         }         public virtual void ShowArea()         {             Console.WriteLine($"Area = {length * breadth} sq. units."); 73

Chapter 4

Know SOLID Principles

            Console.WriteLine("----------");         }     }     class Program     {         static void Main(string[] args)         {             Console.WriteLine("***Understanding LSP.***");             Console.WriteLine("Rectangle-1:");             Rectangle rectangle = new Rectangle(10, 5.5);             rectangle.ShowArea();             Console.WriteLine("Rectangle-2:");             rectangle = new Rectangle(25);             rectangle.ShowArea();             Console.ReadKey();         }     } } When you run this program, you see the following output: ***Understanding LSP.*** Rectangle-1: Length = 10 units. Breadth = 5.5 units. Area = 55 sq. units. ---------Rectangle-2: Length = 25 units. Breadth = 2 units. Area = 50 sq. units. ----------

74

Chapter 4

Know SOLID Principles

The program is simple and the output is straightforward. Let’s suppose you now consider squares. You know that a square can be thought of as a special type of rectangle. So, you override the ShowArea() method inside the Square class and write this additional piece of code:     class Square : Rectangle     {         public Square(double length) :             base(length)         {         }         public override void ShowArea()         {             Console.WriteLine($"Area = {length * length} sq. units.");             Console.WriteLine("----------");         }     } Now you create a square and add it to your old rectangle list: rectangle = new Square(25); rectangle.ShowArea(); Here is the Main() method for your immediate reference (see the changes in bold):   static void Main(string[] args)         {             Console.WriteLine("***Understanding LSP.***");             Console.WriteLine("Rectangle-1:");             Rectangle rectangle = new Rectangle(10, 5.5);             rectangle.ShowArea();             Console.WriteLine("Rectangle-2:");             rectangle = new Rectangle(25);             rectangle.ShowArea();

75

Chapter 4

Know SOLID Principles

            Console.WriteLine("Rectangle-3:");             rectangle = new Square(25);             rectangle.ShowArea();             Console.ReadKey();         } Now run the program again. This time you see the following output: ***Understanding LSP.*** Rectangle-1: Length = 10 units. Breadth = 5.5 units. Area = 55 sq. units. ---------Rectangle-2: Length = 25 units. Breadth = 2 units. Area = 50 sq. units. ---------Rectangle-3: Length = 25 units. Breadth = 2 units. Area = 625 sq. units. ---------Notice the final part of the output. Anyone who sees this output would be confused. Why? We know that special rectangles like squares require a different formula. The incorrect breadth value causes this confusion. The user may believe that he or she is interacting with proper rectangles only. Also, if a third-party team wrote test cases to verify ShowArea() using Assert statements, it would be likely to fail. Why? A third-party tester who saw this might assume traditional rectangles, not a specific type of rectangle. You understand the reason: when you consider a square, its length and breadth are the same, and they change together, which is not the case for a rectangle. In simple words, when you have a traditional rectangle, you need both the length and breadth to calculate the area. But for a square, the length and breadth are the same; so, only one of them is sufficient. In this program, you cannot simply substitute a square with a rectangle and 76

Chapter 4

Know SOLID Principles

vice versa. You need to employ some programming logic when your rectangle is a special type (i.e., when it is a square). I hope you understand the problem with this design!

Note  Is it bad to think, since a square IS-A a rectangle, a Square class should inherit from a Rectangle class? Ideally, the answer is no. The problem is not in the IS-A test; the potential problem is in the word “should.” When you design a Square class, it is not always necessary for you to start from a Rectangle class and inherit from it. In addition, once you learn LSP, you’ll see that some “special problems” are code-specific and you cannot predict everything in advance. Let us suppose you rewrite your Square class in such a way that once a user supplies a length, the breadth takes the same value (and vice versa) and your code works fine for both rectangles and squares now. Even in this case, you may see “unwanted” results. For example, assume that you have a method that takes a Rectangle instance as an argument and changes either the length or the breadth value of this instance. Now, if instead of passing a Rectangle instance, you pass a Square instance, this method will corrupt the Square object because it changes only length or breadth, but not both. We can further fix this problem. But remember that in programming, changes in derived classes should not cause changes in base classes. If that happens by mistake, the program violates the OCP too. Since you do not know everything in advance, it makes sense to go in the reverse direction in some specific scenarios.

Initial Program Let’s consider a better scenario. Assume that you have a payment portal. In this portal, a registered user can raise a payment request. You use the method ProcessNewPayment() for this. In this portal, you can also show a user’s last payment detail. You use the method LoadPreviousPaymentInfo() for this purpose. Here is a sample code segment for this: interface IUser     {         void LoadPreviousPaymentInfo();         void ProcessNewPayment();     } 77

Chapter 4

Know SOLID Principles

    class RegisteredUser : IUser     {         string name = String.Empty;         public RegisteredUser(string name)         {             this.name = name;         }         public void LoadPreviousPaymentInfo()         {             Console.WriteLine($"Welcome {name}. Here is your last payment details.");         }         public void ProcessNewPayment()         {             Console.WriteLine($"Processing {name}'s current payment request.");         }     } Furthermore, let’s assume you create a helper class UserManagementHelper to display all previous payments and new payment requests of these users. You use ShowPreviousPayments() and ProcessNewPayments() for these activities. These methods call the LoadPreviousPaymentInfo() and ProcessNewPayment() methods on the respective IUser instances in the foreach loop. Here is the code segment for this: class UserManagementHelper     {         List users = new List();         public void AddUser(IUser user)         {             users.Add(user);         }         public void ShowPreviousPayments()         {

78

Chapter 4

Know SOLID Principles

            foreach (IUser user in users)             {                 user.LoadPreviousPaymentInfo();                 Console.WriteLine("------");             }         }         public void ProcessNewPayments()         {             foreach (IUser user in users)             {                 user.ProcessNewPayment();                 Console.WriteLine("***********");             }         }     } Inside the client code, you create two users and show their current payment requests along with previous payments. Everything is OK so far.

Demonstration 5 Go through the complete demonstration now. using System; using System.Collections.Generic; namespace WithoutLSPDemo {     interface IUser     {         void LoadPreviousPaymentInfo();         void ProcessNewPayment();     }     class RegisteredUser : IUser     {         string name = String.Empty;

79

Chapter 4

Know SOLID Principles

        public RegisteredUser(string name)         {             this.name = name;         }         public void LoadPreviousPaymentInfo()         {             Console.WriteLine($"Welcome, {name}. Here are your last payment details.");         }         public void ProcessNewPayment()         {             Console.WriteLine($"Processing {name}'s current payment request.");         }     }     class UserManagementHelper     {         List users = new List();         public void AddUser(IUser user)         {             users.Add(user);         }         public void ShowPreviousPayments()         {             foreach (IUser user in users)             {                 user.LoadPreviousPaymentInfo();                 Console.WriteLine("------");             }         }         public void ProcessNewPayments()         {             foreach (IUser user in users)             { 80

Chapter 4

Know SOLID Principles

                user.ProcessNewPayment();                 Console.WriteLine("***********");             }         }     }     class Program     {         static void Main(string[] args)         {             Console.WriteLine("***A demo without LSP.***");             UserManagementHelper helper = new UserManagementHelper();             // Instantiating two registered users             RegisteredUser robin = new RegisteredUser("Robin");             RegisteredUser jack = new RegisteredUser("Jack");             // Adding the users to usermanager             helper.AddUser(robin);             helper.AddUser(jack);             // Processing the payments using             // the helper class.             helper.ShowPreviousPayments();             helper.ProcessNewPayments();             Console.ReadKey();         }     } }

Output Here is the output: ***A demo without LSP.*** Welcome, Robin. Here are your last payment details. -----Welcome, Jack. Here are your last payment details. -----81

Chapter 4

Know SOLID Principles

Processing Robin's current payment request. *********** Processing Jack's current payment request. *********** This program seems to be fine. Now assume that you need to support guest users too. You understand that you can process a guest user’s payment request, but you would not show his last payment detail. So, you create the following class that implements IUser: class GuestUser : IUser     {         string name = String.Empty;         public GuestUser()         {             this.name = "guest user";         }         public void LoadPreviousPaymentInfo()         {             throw new NotImplementedException();         }         public void ProcessNewPayment()         {             Console.WriteLine($"Processing {name}'s current payment request.");         }     } Inside Main(), you create a guest user instance now and try to use your helper class in the same manner. Here is the new client code (notice the changes in bold). For your easy understanding, I have added a comment to draw your attention to the code that causes the problem now. class Program     {         static void Main(string[] args)         { 82

Chapter 4

Know SOLID Principles

            Console.WriteLine("***A demo without LSP.***");             UserManagementHelper helper = new UserManagementHelper();             // Instantiating two registered users             RegisteredUser robin = new RegisteredUser("Robin");             RegisteredUser jack = new RegisteredUser("Jack");             // Adding the users to usermanager             helper.AddUser(robin);             helper.AddUser(jack);             GuestUser guestUser1 = new GuestUser();             helper.AddUser(guestUser1);             // Processing the payments using             // the helper class.             // You can see the problem now.             helper.ShowPreviousPayments();             helper.ProcessNewPayments();             Console.ReadKey();         }     } This time, you get a surprise and encounter an exception. See Figure 4-1.

Figure 4-1.  The program encounters the NotImplementedException

83

Chapter 4

Know SOLID Principles

Though GuestUser implements IUser, it causes UserManagementHelper to break. The following loop: foreach (IUser user in registeredUsers) {     user.LoadPreviousPaymentInfo();     Console.WriteLine("------"); } causes this trouble. In every iteration, you called the method LoadPreviousPaymentInfo() on the respective IUser object, and the exception was raised for the GuestUser instance. The previous working solution does not work now because the GuestUser violates the LSP. What is the solution? Go to the next section.

Better Program The first obvious solution that may come into your mind is to employ an if-else chain to verify whether the IUser instance is a GuestUser or a RegisteredUser. It is a bad solution because if you have another special type of user, you again verify it inside the if-else chain. Most important, you violate the OCP each time you modify the existing class using this if-else chain. So, let us search for a better solution. In the upcoming program, I remove the ProcessNewPayment() method from the IUser interface. I place this method into another interface, INewPayment. As a result, now I have two interfaces with the specific operations. Since all types of users can raise a new payment request, the concrete classes RegisteredUser and GuestUser both implement the INewPayment interface. But you show the last payment detail for registered users only. So, the RegisteredUser class implements the IUser interface. I always advocate for a proper name. Since IUser contains the LoadPreviousPaymentInfo() method, it makes sense that you choose a better name, say IPreviousPayment instead of IUser. I have adjusted these new names in the helper class too.

Demonstration 6 Here is the modified implementation. using System; using System.Collections.Generic; 84

Chapter 4

Know SOLID Principles

namespace LSPDemo {     interface IPreviousPayment     {         void LoadPreviousPaymentInfo();     }     interface INewPayment     {         void ProcessNewPayment();     }     class RegisteredUser : IPreviousPayment, INewPayment     {         string name = String.Empty;         public RegisteredUser(string name)         {             this.name = name;         }         public void LoadPreviousPaymentInfo()         {             Console.WriteLine($"Welcome, {name}. Here are your last payment details.");         }         public void ProcessNewPayment()         {             Console.WriteLine($"Processing {name}'s current payment request.");         }     }     class GuestUser : INewPayment     {         string name = String.Empty;         public GuestUser()         {             this.name = "guest user";         } 85

Chapter 4

Know SOLID Principles

        public void ProcessNewPayment()         {             Console.WriteLine($"Processing a {name}'s current payment request.");         }     }     class UserManagementHelper     {         List previousPayments = new List();         List newPaymentRequests = new List();         public void AddPreviousPayment(IPreviousPayment previousPayment)         {             previousPayments.Add(previousPayment);         }         public void AddNewPayment(INewPayment newPaymentRequest)         {             newPaymentRequests.Add(newPaymentRequest);         }         public void ShowPreviousPayments()         {             foreach (IPreviousPayment user in previousPayments)             {                 user.LoadPreviousPaymentInfo();                 Console.WriteLine("------");             }         }         public void ProcessNewPayments()         {             foreach (INewPayment payment in newPaymentRequests)             {                 payment.ProcessNewPayment();                 Console.WriteLine("***********");             } 86

Chapter 4

Know SOLID Principles

        }     }     class Program     {         static void Main(string[] args)         {             Console.WriteLine("***A demo that follows LSP.***");             UserManagementHelper helper = new UserManagementHelper();             // Instantiating two registered users.             RegisteredUser robin = new RegisteredUser("Robin");             RegisteredUser jack = new RegisteredUser("Jack");             // Adding the info to helper.             helper.AddPreviousPayment(robin);             helper.AddPreviousPayment(jack);             helper.AddNewPayment(robin);             helper.AddNewPayment(jack);             // Instantiating a guest user.             GuestUser guestUser1 = new GuestUser();             helper.AddNewPayment(guestUser1);             // Retrieve all the previous payments             // of registered users.             helper.ShowPreviousPayments();             // Process all new payment requests             // from all users.             helper.ProcessNewPayments();             Console.ReadKey();         }     } }

87

Chapter 4

Know SOLID Principles

Output Here is the output: ***A demo that follows LSP.*** Welcome, Robin. Here are your last payment details. -----Welcome, Jack. Here are your last payment details. -----Processing Robin's current payment request. *********** Processing Jack's current payment request. *********** Processing a guest user's current payment request. ***********

Analysis What are the key changes? Notice that in Demonstration 5, ShowPreviousPayments() and ProcessNewPayments() both accepted IUser instances as arguments. Now ShowPreviousPayments() accepts IPreviousPayment instances and ProcessNewPayments() accepts INewPayment instances as arguments. This new structure solves the problem that we faced in Demonstration 5.

Interface Segregation Principle (ISP) One often sees fat interfaces that contain many methods. A class that implements such an interface may not need all those methods. So, why does the interface contain all those methods? The simple answer is: to support some of the implementing classes of this interface. This is the area the interface segregation principle (ISP) focuses on. It suggests to not pollute an interface with unnecessary methods to only support one (or some) of the implementing classes of the interface. The idea is: A client should not depend on a method that it does not use. Once you understand this principle, you’ll realize that I already used ISP when I showed you a better design following LSP. For now, let us consider an example with our full focus on ISP. 88

Chapter 4

Know SOLID Principles

POINTS TO REMEMBER Note the following points before you proceed further: • A client means any class that uses another class (or interface). • The word “interface” of the interface segregation principle (ISP) is not limited to a C# interface. The same concept applies to any base class interface, such as an abstract class or a simple base class. • Many examples across different sources explain the violation of ISP with an emphasis on throwing a NotImplementedException(). In Demonstration 7, I also demonstrate to you such an example. It helps me to show you the disadvantages of an approach that does not follow ISP (or LSP). You saw earlier that LSP can deal with this kind of problem. • ISP suggests your class should not depend on interface methods that it does not use. This statement will make sense to you when you go through the following example.

I nitial Program Assume that you have an interface IPrinter with two methods—say, PrintDocument() and SendFax(). There are several users of this class. For simplicity, let’s call them BasicPrinter and AdvancedPrinter. Figure 4-2 shows a simple class diagram for this.

Figure 4-2.  The IPrinter class hierarchy

89

Chapter 4

Know SOLID Principles

A basic printer can print some documents. It does not support any other functionality. So, the BasicPrinter needs the PrintDocument() method only. An advanced printer can print documents as well as send faxes. So, the AdvancedPrinter needs both these methods. In this case, if there is a change to SendFax() in the interface IPrinter, it will force the BasicPrinter code to recompile. This situation is unwanted and can cause potential problems for you in the future. You have seen such a problematic situation in Demonstration 5. Later, I showed you a solution in Demonstration 6 where I segregated the interface IUser into IPreviousPayment and INewPayment. This time, I followed the ISP. This principle suggests you design your interface with the proper methods that a particular client may need. Now you ask me: Why does a user invite the problem in the first place? Or, why does a user need to change a base class (or an interface)? To answer this, assume that you want to show which type of fax you are using to send. We know the different variations of fax methods, such as LanFax, InternetFax (or EFax), and AnalogFax. So, earlier, the SendFax() method did not use any parameters, but now it needs to accept a parameter to show the type of fax it uses. To demonstrate this further, let us suppose you have a fax hierarchy that may look like the following: interface IFax   {       void FaxType();   }   class LanFax : IFax   {       public void FaxType()       {           Console.WriteLine("Using lanfax to send the fax.");       }   }   class EFax : IFax   {       public void FaxType()       {           Console.WriteLine("Using internet fax(efax) to send the fax.");       }   } 90

Chapter 4

Know SOLID Principles

To use this inheritance chain, let us assume you update the original SendFax() in AdvancedPrinter to SendFax(IFax faxType), which requires you to change the interface IPrinter. When you do this, you need to update the BasicPrinter class too to accommodate this change. Now you see the problem!

Note  You can see that a change in AdvancedPrinter causes changes in the interface IPrinter, which in turn causes BasicPrinter to update its fax method. Though BasicPrinter does not need this fax method at all, a change in this method in another client of IPrinter forces BasicPrinter to change and recompile. ISP suggests you take care of this kind of scenario. So, when you see a fat interface, ask yourself whether all those interface methods are required for each client. If not, split it into smaller interfaces that are relevant to specific clients. If you understood the previous discussion, you’ll understand why do I not suggest you write the following code: interface IPrinter     {         void PrintDocument();         void SendFax();     } You can see that IPrinter contains both the PrintDocument() and the SendFax() methods. If you start your coding by considering advanced printers that can print as well as send a fax, it is fine. But in a later stage, if your program needs to support basic printers too, you will be forced to write something like this: class BasicPrinter : IPrinter {     public void PrintDocument()     {         Console.WriteLine("A basic printer can print documents.");     }

91

Chapter 4

Know SOLID Principles

    public void SendFax()     {         throw new NotImplementedException();     } } And this code can cause a potential problem for you! You understand that a basic printer cannot send a fax. But since BasicPrinter implements IPrinter, it needs to provide a SendFax() implementation. As a result, when SendFax() changes in the IPrinter interface, the BasicPrinter needs to accommodate the change. The ISP suggests you avoid this kind of situation.

Note  In this context, can you remember the issue in Demonstration 5? When you throw the exception and try to use polymorphic code, you see the impact of violating the LSP. Once you modify IPrinter, you violate OCP too. In this case, inside Main(), you cannot write polymorphic code like the following (because the last line of this code segment will throw a run-time error): IPrinter printer = new AdvancedPrinter(); printer.PrintDocument(); printer.SendFax(); printer = new BasicPrinter(); printer.PrintDocument(); printer.SendFax();// Will throw error And, you cannot write something like: List printers = new List                          {                           new AdvancedPrinter(),                           new BasicPrinter()                           }; foreach( IPrinter device in printers)   {     device.PrintDocument(); 92

Chapter 4

Know SOLID Principles

    //device.SendFax(); // Will throw error   } In both these cases, you will see run-time exceptions.

Demonstration 7 Here is the complete demonstration that does not follow ISP: using System; using System.Collections.Generic; namespace WithoutISPDemo {     interface IPrinter     {         void PrintDocument();         void SendFax();     }     class BasicPrinter : IPrinter     {         public void PrintDocument()         {             Console.WriteLine("A basic printer can print documents.");         }         public void SendFax()         {             throw new NotImplementedException();         }     }     class AdvancedPrinter : IPrinter     {         public void PrintDocument()         {             Console.WriteLine("An advanced printer can print documents.");         } 93

Chapter 4

Know SOLID Principles

        public void SendFax()         {             Console.WriteLine("An advanced printer can send a fax.");         }     }     class Program     {         static void Main(string[] args)         {             Console.WriteLine("***A demo without ISP.***");             IPrinter printer = new AdvancedPrinter();             printer.PrintDocument();             printer.SendFax();             printer = new BasicPrinter();             printer.PrintDocument();             //printer.SendFax();//Will throw error             Console.ReadKey();         }     } }

Output Here is the output: ***A demo without ISP.*** An advanced printer can print documents. An advanced printer can send a fax. A basic printer can print documents.

Analysis You can see that to prevent the run-time exceptions, I needed to comment out a line of code. I kept this dead code for this discussion. You already know that you should avoid this kind of commented code. Most important, as said before, in this design, if you 94

Chapter 4

Know SOLID Principles

change the SendFax() method’s signature in AdvancedPrinter, you need to make the change in IPrinter, which causes BasicPrinter to change and recompile. Think about the problem from another angle too. Assume that you need to support another printer that can print, fax, and photocopy. In this case, if you add a photocopying method in the IPrinter interface, both the existing clients—BasicPrinter and AdvancedPrinter— need to accommodate the change.

Better Program Let’s find a better solution. You know that there are two different activities: one is to print some document and the other one is to send a fax. So, in the upcoming example, I create two interfaces: IPrinter and IFaxDevice. IPrinter contains the PrintDocument() method and IFaxDevice contains the SendFax() method. The idea is simple: •

The class that wants the print functionality implements the IPrinter interface, and the class that wants the fax functionality implements the IFaxDevice interface.



If a class wants both these functionalities, it implements both interfaces.

Note  You should not assume that ISP says an interface should have only one method. In my example, there are two methods in the IPrinter interface, and the BasicPrinter class needs only one of them. That is the reason you see segregated interfaces with a single method only.

Demonstration 8 Here is the complete implementation: using System; namespace ISPDemo {     interface IPrinter     { 95

Chapter 4

Know SOLID Principles

        void PrintDocument();     }     interface IFaxDevice     {         void SendFax();     }     class BasicPrinter : IPrinter     {         public void PrintDocument()         {             Console.WriteLine("A basic printer can print documents.");         }     }     class AdvancedPrinter : IPrinter, IFaxDevice     {         public void PrintDocument()         {             Console.WriteLine("An advanced printer can print documents.");         }         public void SendFax()         {             Console.WriteLine("An advanced printer can send a fax.");         }     }     class Program     {         static void Main(string[] args)         {             Console.WriteLine("***A demo that follows ISP.***");             IPrinter printer = new BasicPrinter();             printer.PrintDocument();             printer = new AdvancedPrinter();             printer.PrintDocument();

96

Chapter 4

Know SOLID Principles

            IFaxDevice faxDevice = new AdvancedPrinter();             faxDevice.SendFax();             Console.ReadKey();         }     } }

Output Here is the output: ***A demo that follows ISP.*** A basic printer can print documents. An advanced printer can print documents. An advanced printer can send a fax.

Analysis You may think that you can provide a default implementation in the interface (C# 8.0 onward, this feature is supported) or an abstract class so that a subclass (or interface-­ implementing class) does not need to worry about the functionality. If you think like this, I remind you that each time you add a method in the base class (or interface), the method needs to be implemented (or available for use) in the derived classes. This kind of practice can violate OCP and LSP, which in turn causes difficult maintenance and reusability issues. For example, if you provide a default fax method in an interface (or an abstract class), BasicPrinter must override it by saying something similar to the following: public void SendFax() {   throw new NotImplementedException(); } And you saw the potential problem with this!

97

Chapter 4

Know SOLID Principles

Note  There is an alternative technique for implementing the ISP: the “delegation” technique. However, delegations increase the run-time (it can be small, but it is nonzero for sure) of an application, which can affect the performance of the application. Also, based on a particular design, a delegated call can create some additional objects. Too many of these objects can cause memory issues, particularly when you deal with very little memory in your application.

Dependency Inversion Principle (DIP) The dependency inversion principle (DIP) involves two important things: •

A high-level concrete class should not depend on a low-level concrete class. Instead, both should depend on abstractions.



Abstractions should not depend upon details. Instead, the details should depend upon abstractions.

We’ll examine both these points. The reason for the first point is simple. If the low-level class changes, the high-level class may need to adjust to the change; otherwise, the application breaks. What does it mean? It says that you should avoid creating a concrete low-level class inside a high-level class. Instead, you should use abstract classes or interfaces. As a result, you remove the tight coupling between the classes. The second point is also easy to understand when you analyze the case study that I discussed in ISP. You saw that if an interface needs to change to support one of its clients, other clients can be impacted due to the change. No client likes to see such an application. So, in your application, if your high-level modules are independent of low-level modules, you can reuse them easily. This idea also helps you design nice frameworks.

98

Chapter 4

Know SOLID Principles

Note  In his book Agile Principles, Patterns and Practices in C#, Robert C. Martin explains that a traditional software development model in those days (such as structured analysis and design) tended to create software where high-level modules depended on low-level modules. But in OOP, a well-designed program opposes the idea. It inverts the dependency structure that often results from a traditional procedural method. This is the reason he used the word “inversion” in this principle.

Initial Program Assume that you have a two-layer application. Using this application, a user can save an employee ID in a database. To demonstrate this, I use a console application instead of a Windows Forms application. Here, you see two classes: UserInterface and OracleDatabase. As per their names, the UserInterface represents a user interface (such as a form where a user can type an employee ID and click the Save button to save the ID in a database). Similarly, the OracleDatabase is used to mimic an Oracle database. Again, for simplicity, there is no actual database in this application, and there is no code to validate an employee ID. Here our focus is on DIP only, so those discussions are not important. Assume that using the SaveEmployeeId() method of the UserInterface, you can save an employee ID to a database. Notice the following code segment inside the UserInterface class: public UserInterface() { this.oracleDatabase = new OracleDatabase(); } public void SaveEmployeeId(string empId) {   // Assume that it is a valid data.   // So, I store it to the database.   oracleDatabase.SaveEmpIdInDatabase(empId); } 99

Chapter 4

Know SOLID Principles

You can see that I instantiate an OracleDatabase object inside the UserInterface constructor. Later, I use this object to invoke the SaveEmpIdInDatabase() method, which does the actual saving inside the Oracle database. The following class diagram (Figure 4-3) shows the high-level class (UserInterface) dependency on the low-level class (OracleDatabase).

Figure 4-3.  A high-level class UserInterface depends on a low-level class OracleDatabase This style of coding is very common. But there are some problems. We’ll discuss them in the analysis section before I show you a better approach.

D  emonstration 9 For now, see the complete program, which does not follow the DIP. using System; namespace WithoutDIPDemo {     class UserInterface     {         readonly OracleDatabase oracleDatabase;         public UserInterface()         {           this.oracleDatabase = new OracleDatabase();         }

100

Chapter 4

Know SOLID Principles

        public void SaveEmployeeId(string empId)         {             // Assuming that this is a valid data.             // So, storing it to the database.             oracleDatabase.SaveEmpIdInDatabase(empId);         }     }     class OracleDatabase     {         public void SaveEmpIdInDatabase(string empId)         {           Console.WriteLine($"The id: {empId} is saved in the oracle database.");         }     } class Program { static void Main(string[] args)   {     Console.WriteLine("***A demo without DIP.***");     UserInterface userInterface = new UserInterface();     userInterface.SaveEmployeeId("E001");     Console.ReadKey();    }   } }

Output Here is the output: ***A demo without DIP.*** The id: E001 is saved in the oracle database.

101

Chapter 4

Know SOLID Principles

A  nalysis The program is simple, but it suffers from the following issues: •

The top-level class (UserInterface) has too much dependency on the bottom-level class (OracleDatabase). These two classes are tightly coupled. So, in the future, if the OracleDatabase class changes, you may need to adjust the changes in the UserInterface.



The low-level class should be available before you write the top-level class. So, you are forced to complete the low-level class before you write or test the high-level class.



What do you do if you use a different database? For example, you may switch from the Oracle database to a MySQL database; or, you may need to support both.

B  etter Program In this program, you see the following hierarchy: interface IDatabase {   void SaveEmpIdInDatabase(string empId); } class OracleDatabase : IDatabase { public void SaveEmpIdInDatabase(string empId) {   Console.WriteLine($"The id: {empId} is saved in the Oracle database.");   } } The first part of DIP suggests we focus on abstraction. This makes the program efficient. So, this time the UserInterface class targets the abstraction IDatabase, instead of a concrete implementation such as OracleDatabase. This gives you the flexibility to consider a new database, such as MYSQLDatabase, as well. The class diagram in Figure 4-­4 describes the scenario. 102

Chapter 4

Know SOLID Principles

Figure 4-4.  A high-level class UserInterface depends on the abstraction IDatabase The second part of the DIP suggests making the IDatabase interface consider the needs of the UserInterface class. This is important because if an interface needs to change to support one of its clients, other clients can be impacted. This is why you should not design IDatabase(abstraction) depending on the needs of OracleDatabase or MySQLDatabase(details).

Demonstration 10 Here is the complete program for you: using System; namespace DIPDemo {     class UserInterface     {         readonly IDatabase database;         public UserInterface(IDatabase database)         {             this.database = database;         } 103

Chapter 4

Know SOLID Principles

        public void SaveEmployeeId(string empId)         {             database.SaveEmpIdInDatabase(empId);         }     }     interface IDatabase     {         void SaveEmpIdInDatabase(string empId);     }     class OracleDatabase : IDatabase     {         public void SaveEmpIdInDatabase(string empId)         {             Console.WriteLine($"The id: {empId} is saved in the Oracle database.");         }     }     class MySQLDatabase : IDatabase     {         public void SaveEmpIdInDatabase(string empId)         {             Console.WriteLine($"The id: {empId} is saved in the MySQL database.");         }     }     class Program     {         static void Main(string[] args)         {             Console.WriteLine("***A demo that follows DIP.***");             //Using Oracle now             IDatabase database = new OracleDatabase();

104

Chapter 4

Know SOLID Principles

            UserInterface userInterface = new UserInterface(database);             userInterface.SaveEmployeeId("E001");             //Using MySQL now             database = new MySQLDatabase();             userInterface = new UserInterface(database);             userInterface.SaveEmployeeId("E002");             Console.ReadKey();         }     } }

Output Here is the output: ***A demo that follows DIP.*** The id: E001 is saved in the Oracle database. The id: E002 is saved in the MySQL database.

Analysis You can see that by following this program, you can resolve all the potential issues of the previous program in Demonstration 9. In short, in OOP, I’d suggest following the following Robert C. Martin quote:

High-level modules simply should not depend on low-level modules in any way. So, when you have a base class and a derived class, your base class should not know about any of its derived classes. But there are few exceptions to this suggestion. For ­example, consider the case where your base class needs to restrict the count of the derived class instances at a certain point. One last point: You can see that in Demonstration 10, the UserInterface class constructor accepts a database parameter. You can provide an additional facility to the user when you use both the constructor and the property inside this class. Here is

105

Chapter 4

Know SOLID Principles

a sample code for you. To follow the naming convention, I needed to choose the name “Database” instead of “database.” I also kept the commented code, so that you can compare it with the existing code. class UserInterface { //readonly IDatabase database; public IDatabase Database { get; set; } public UserInterface(IDatabase database) {   //this.database = database;   this.Database = database; } public void SaveEmployeeId(string empId) {   //database.SaveEmpIdInDatabase(empId);   Database.SaveEmpIdInDatabase(empId); } } What is the benefit of doing this? Now you can instantiate a database while instantiating the UserInterface class and change it later using the Database property. Here is a sample code for you that you can append to test at the end of Main(): // Additional code for demonstration purpose userInterface.Database = new OracleDatabase(); userInterface.SaveEmployeeId("E003"); You can follow the same technique for similar examples that are used in this book.

Summary The SOLID principles are the fundamental guidelines of object-oriented design. These high-level concepts can help you make better software. They are neither rules nor laws, but they can help you think of possible scenarios/outcomes in advance. In this chapter, I showed you applications that follow (and do not follow) these principles and discussed the pros and cons. Let’s have a quick review. 106

Chapter 4

Know SOLID Principles

SRP says that a class should have only one reason to change. Using SRP you can write cleaner and less fragile code. You identify the responsibilities and make classes based on each responsibility. What is a responsibility? It is a reason for a change. But you should not assume that a class should have a single method only. If multiple methods help you to achieve a single responsibility, your class can contain all these methods. You are OK to bend this rule based on the nature of possible changes. The reason for this is that if you have too many classes in an application, it is difficult to maintain. But when you know this principle and think carefully before you implement a design, you can avoid those typical mistakes that I discussed earlier. Robert C. Martin mentioned the OCP as the most important object-oriented design principle. The OCP suggests that software entities (a class, module, method, etc.) should be open for extension, but closed for modification. If you do not touch a running code, you do not break it. For new features, you add new codes, but do not disturb the existing code. It helps you to save time rather than testing the entire workflow again. Instead, you focus on the newly added code and test that part. This principle is often hard to achieve, but even partial OCP compliance can provide a benefit to you in the long run. In many cases, when you violate the OCP, you break the SRP too. The idea of LSP is that you should be able to substitute a parent (or base) type with a subtype. It is your responsibility to write true polymorphic code using the LSP. This principle is very important and was discussed with two different case studies. Using this principle, you can avoid the long tail of if-else chains and make your code OCP compliant too. The idea behind ISP is that a client should not depend on a method that it does not use. This is why you may need to split a fat interface into multiple interfaces. I have shown you a simple technique to implement the idea. When you do not modify an existing interface (or an abstract class or a simple base class), you follow OCP. When you do not throw the NotImplementedException(), you do not break LSP. This is why an ISP-compliant application can help you make OCP- and LSP-compliant applications too. You can make an ISP-compliant application using the delegation technique, which I did not discuss in this book. But the important point is that when you use delegation, you increase the run-time (you may say it is negligible, but it is nonzero for sure), which can affect a time-sensitive application. Using delegation, you may create a new object(s) when a client uses the application. However, this may cause memory issues in certain scenarios.

107

Chapter 4

Know SOLID Principles

DIP suggests two important points for us. First, a high-level concrete class should not depend on a low-level concrete class. Instead, both should depend on abstractions. Second, the abstractions should not depend upon details. Instead, the details should depend upon the abstractions. When you follow the first part, your application will be efficient and flexible; you can consider new concrete implementations in your application. When you analyze the second part of this principle, you will understand that you should not change an existing base class or interface to support one of its clients. This can cause another client to break, and you will have violated the OCP in such a case. I analyzed the importance of both these points.

108

CHAPTER 5

Use the DRY Principle The Don’t Repeat Yourself (DRY) principle is another important principle that a professional coder must follow when he writes a piece of code for an application. The Single Responsibility Principle (SRP) and Open/Closed Principle (OCP) principles that you learned in Chapter 4 are related to the DRY principle. We came to know about this principle from Andy Hunt and Dave Thomas in their famous book, The Pragmatic Programmer. In this book, the DRY principle is stated as follows:

Every piece of knowledge must have a single, unambiguous, authoritative representation within a system. It may seem complicated when you read this for the first time. The goal of this chapter is to help you understand this principle with a case study.

R  easons for DRY Code duplication can cause an application to fail. Programmers often call such instances evils in the software. Now the question is—why do we see duplicated code? There are a variety of reasons. Let’s examine some of them with examples: •

A programmer cannot resist doing a simple copy/paste, which appears to him to be the shortest path to success.



The project deadline is approaching. The developer assumes that a certain number of duplicates are OK at this moment. He plans to remove these duplicates in the next release, but he forgets this later.



Duplicates are seen in code comments too. Consider an example: a developer knows the code very well. He does not need the documentation to understand the logic of the code. A new requirement forces him to update a portion of the code. So, he copies and pastes an

© Vaskaran Sarcar 2021 V. Sarcar, Simple and Efficient Programming with C#, https://doi.org/10.1007/978-1-4842-7322-7_5

109

Chapter 5

Use the DRY Principle

existing piece of code with existing comments and starts working on the code segment. Once the update is done, due to various reasons, he forgets to update the associated comments. •

A tester may need to pass the same input to verify various methods in a test suite.



Sometimes duplicates are hard to avoid. Project standards may require a developer to put duplicate information in the code. Take another example: suppose your software targets multiple platforms that use different programming languages and/or development environments. In this case, you may need to duplicate shared information (such as methods).



Apart from this, a programming language can have a structure that itself duplicates some information.

And so on. In computer science, many principles are followed to avoid code duplications. For example, the database normalization techniques try to eliminate duplicate data. In Chapter 2, you saw that you can put a common method in an abstract base class to avoid duplicating the method in the derived classes. I must say that finding duplicate codes is not always easy. For example, consider the following code segment (Code Segment 1), which has two methods. Code Segment 1 public void DisplayCost() { Console.WriteLine("Game name: SuperGame"); Console.WriteLine("Version:1.0"); Console.WriteLine("Actual cost is:$1000"); } public void DisplayCostAfterDiscount() { Console.WriteLine("Game name: SuperGame"); Console.WriteLine("Version:1.0"); Console.WriteLine("The discounted price for festive season is:$800"); } 110

Chapter 5

Use the DRY Principle

You can easily see that the initial two lines are common to both methods. But detection of duplicates is not as straightforward if they are intermixed with other code or comments. For example, consider Code Segment 2. Code Segment 2 public void DisplayCost() { Console.WriteLine("\AbcCompany SuperGame's price details:"); Console.WriteLine("Version:1.0 cost is:$1000"); } public void DisplayCostAfterDiscount() { Console.WriteLine("\AbcCompany offers festive season discount."); Console.WriteLine("Discounted price detail:"); Console.WriteLine("Game: SuperGame. Version: 1.0. Discounted price:$800"); } On careful observation, you can see that in both code segments, the company name, game name, and version detail of the software are repeated. Though in the first code segment it is very easy to find the duplicate codes, in the second code segment you need to read the code carefully. These code segments contain only two methods. In a real-world application, there are numerous methods, and not all the methods are present in the same file. So, if you spread duplicate information across the files, a simple update can cause the software to show inconsistent behavior. During an update operation, if you have n number of duplicates, you need n-fold modifications, and you cannot miss any of them. This is why you need to be careful about them. Violation of the DRY principle causes you to get a WET solution. It commonly stands for “Write Every Time,” “Write Everything Twice,” “We Enjoy Typing,” or “Waste Everyone's Time.” Like in the previous chapters, I will start with a program that seems fine at the beginning. We will analyze this program and make it better by eliminating the redundant code. You can follow the same approach when you confront a similar situation.

111

Chapter 5

Use the DRY Principle

Initial Program Here is a simplistic example for you. The key assumptions in this program are as follows: •

There is game software called SuperGame. You create a class to represent this game.



The AboutGame() method tells some useful information about this software. For example, it states that the minimum age for using this software is 10. It also shows the current version of the game.



DisplayCost() shows the price of the latest version of this software.



A buyer can get up to a 20% discount. The DisplayCostAfterDiscount() shows the discounted price of the latest software.

Demonstration 1 Assume that someone has written the following program. It compiles and runs successfully. Let’s see the output and then follow the analysis section. using System; namespace WithoutDRYDemo { class SuperGame { public void AboutGame() { Console.WriteLine("Game name: SuperGame"); Console.WriteLine("Minimum age: 10 years and above."); Console.WriteLine("Current version: 1.0."); Console.WriteLine("It is a AbcCompany product."); }

112

Chapter 5

Use the DRY Principle

public void DisplayCost() { Console.WriteLine("\AbcCompany SuperGame's price details:"); Console.WriteLine("Version:1.0 cost is:$1000"); } public void DisplayCostAfterDiscount() { Console.WriteLine("\n AbcCompany offers a festive season discount."); Console.WriteLine("Discounted price detail:"); Console.WriteLine("Game: SuperGame. Version: 1.0 Discounted price:$800"); } } class Program {   static void Main()    {     Console.WriteLine("***A demo without DRY principle.***");     SuperGame superGame = new SuperGame();     superGame.AboutGame();     superGame.DisplayCost();     superGame.DisplayCostAfterDiscount();     Console.ReadKey();     }   } }

Output Here is the output: ***A demo without DRY principle.*** Game name: SuperGame Minimum age: 10 years and above. Current version: 1.0. It is a AbcCompany product. 113

Chapter 5

Use the DRY Principle

AbcCompany SuperGame's price details: Version:1.0 cost is:$1000 AbcCompany offers a festive season discount. Discounted price detail: Game: SuperGame. Version: 1.0 Discounted price:$800

Analysis Can you see the problems with this program? How many times do you see the company name AbcCompany and the version detail? I know that it’s a simple program, but consider the case I mentioned before. These methods can appear in different modules, and in such a case if you needed to update the company info or the version detail, you would need to figure out all the places it is used before you could provide the update. This is where the DRY principle comes in. This program suffers from the use of hard-coded strings. The solution to this problem is straightforward. You can contain those strings that appear in multiple places in a single location. Then you share this code segment with other parts of the program. As a result, when you update a string in the shared location, the change reflects properly in every place it appears. So, the basic idea is that if you see a common code in multiple locations, you separate the common parts from the remaining parts, put them in a single location, and call this common code from other parts of the program. In this way, you avoid the copy/ paste technique, which may seem to be easy and appealing at the beginning. The Once and Only Once principle is similar to DRY. You can apply this principle to a functional behavior, and you can think of it as a subset of DRY. In simple words, the core ideas of DRY and Once and Only Once are the same. Can you make Demonstration 1 better? Let us look at the following program.

Better Program This program uses a constructor to initialize the values. You can use these values in the instance methods of the class.

114

Chapter 5

Use the DRY Principle

Demonstration 2 Here is an improved version of Demonstration 1: using System; namespace ImprovedVersion {     class SuperGame     {         readonly string companyName;         readonly string gameName;         readonly double minimumAge;         readonly string version;         readonly double actualCost;         readonly double discountedCost;         public SuperGame()         {             companyName = "AbcCompany";             gameName = "SuperGame";             version = "1.0";             minimumAge = 10;             actualCost = 1000;             discountedCost = 800;         }         public void AboutGame()         {          Console.WriteLine($"Game name: {gameName}");          Console.WriteLine($"Minimum age: {minimumAge} years and above.");          Console.WriteLine($"Current version: {version}.");          Console.WriteLine($"It is a {companyName} product.");         }         public void DisplayCost()         {          Console.WriteLine($"\n{companyName} SuperGame's price details:");          Console.WriteLine($"Version:{version} " +                            $"cost is: {actualCost}");         } 115

Chapter 5

Use the DRY Principle

        public void DisplayCostAfterDiscount()         {          Console.WriteLine($"\n{companyName} offers festive season discount.");         Console.WriteLine("Discounted price detail:");         Console.WriteLine($"Game: {gameName}. " +                 $"Version: {version}. " +                 $"Discounted price:{discountedCost}");         }     }     class Program     {         static void Main()         {             Console.WriteLine("***An improved version using DRY principle.***");             SuperGame superGame = new SuperGame();             superGame.AboutGame();             superGame.DisplayCost();             superGame.DisplayCostAfterDiscount();             Console.ReadKey();         }     } }

Output Here is the output of this program: ***An improved version using DRY principle.*** Game name: SuperGame Minimum age: 10 years and above. Current version: 1.0. It is a AbcCompany product. AbcCompany SuperGame's price details: Version:1.0 cost is:1000 116

Chapter 5

Use the DRY Principle

AbcCompany offers a festive season discount. Discounted price detail: Game: SuperGame. Version: 1.0. Discounted price:800 You can see that this program produces the same output except for the first line, which prints that it is an improved version.

A  nalysis Still, there is some repetition in this example. Notice that the company name is shown in AboutGame(),DisplayCost(), and DisplayCostAfterDiscount(). This is alright here because I want a client to display the company name in any of the methods. But you can improve this program. The initial version of the software and the company name may not change for a different game (made by the same company), but the name of the game and price detail are likely to change. So, I want to improve the program logic in these areas. In addition, if you became familiar with the SOLID principles in the previous chapter (Chapter 4), you know that this program does not follow the SRP. In short, you may need to update this program in the future for various reasons, such as the following: •

The cost of the software can be changed.



The discounted price can be changed.



The version detail can be changed.



The name of the game can be changed.



Also, the company name itself can be changed.

So, I move the company name, game name, version, and age requirement into a new class, GameInfo. The actual price and the discounted price are moved into a different class, GamePrice. In addition, this time I use properties, so you can apply changes to the initial values in a later stage using these properties. In this upcoming program, when you instantiate a GameInfo instance, you supply the name of the game, but before that, you initialize a GameInfo instance and a GamePrice instance. This activity helps you to instantiate a game instance with the default information stored in GameInfo and GamePrice. As said before, you can change these values using various properties of these classes.

117

Chapter 5

Use the DRY Principle

Now, go through a proposed improvement. You can follow a similar structure to incorporate changes with minimum effort.

Demonstration 3 Here is an improved version of Demonstration 2: using System; namespace DRYDemo {     class GameInfo     {         public string CompanyName { get; set; }         public string GameName { get; set; }         public string Version { get; set; }         public double MinimumAge { get; set; }         public GameInfo(string gameName)         {             CompanyName = "AbcCompany";             GameName = gameName;             Version = "1.0";             MinimumAge = 10.5;         }     }     class GamePrice     {         public double Cost { get; set; }         public double DiscountedCost { get; set; }         public GamePrice()         {             Cost = 1000;             DiscountedCost = 800;         }     } 118

Chapter 5

Use the DRY Principle

    class Game     {         readonly string companyName;         readonly string gameName;         readonly double minimumAge;         readonly string version;         readonly double actualCost;         readonly double discountedCost;         public Game(                 GameInfo gameInfo,                 GamePrice gamePrice                 )             {             companyName = gameInfo.CompanyName;             gameName = gameInfo.GameName;             version = gameInfo.Version;             minimumAge = gameInfo.MinimumAge;             actualCost = gamePrice.Cost;             discountedCost = gamePrice.DiscountedCost;             }         public void AboutGame()         {             Console.WriteLine($"Game name: {gameName}");             Console.WriteLine($"Minimum age: {minimumAge} years and above.");             Console.WriteLine($"Current version: {version}.");             Console.WriteLine($"It is a {companyName} product.");         }         public void DisplayCost()         {             Console.WriteLine($"\n{companyName} {gameName}'s price details:");             Console.WriteLine($"Version:{version},cost:${actualCost}");         }

119

Chapter 5

Use the DRY Principle

        public void DisplayCostAfterDiscount()         {             Console.WriteLine($"\n{companyName} offers a festive season discount.");             Console.WriteLine("Discounted price detail:");             Console.WriteLine($"Game: {gameName},Version: {version},Discounted price:${discountedCost}");         }     }     class Program     {         static void Main(string[] args)         {         Console.WriteLine("*** Another improved version following the DRY principle. ***");         // Initial setup         GameInfo gameInfo = new GameInfo("SuperGame");         GamePrice gamePrice = new GamePrice();         // Create the game instance with default setup         Game game = new Game(gameInfo, gamePrice);         // Display the default game detail.         game.AboutGame();         game.DisplayCost();         game.DisplayCostAfterDiscount();         Console.WriteLine("------------");         Console.WriteLine("Changing the game version and price now.");         // Changing some of the game info         gameInfo.Version = "2.0";         gameInfo.MinimumAge = 9.5;         // Changing the game cost         gamePrice.Cost = 1500;         gamePrice.DiscountedCost = 1200; 120

Chapter 5

Use the DRY Principle

       // Updating the game instance         game = new Game(gameInfo, gamePrice);        // Display the latest detail         game.AboutGame();         game.DisplayCost();         game.DisplayCostAfterDiscount();         Console.ReadKey();         }     } }

Output Here is the new output that reflects the changes in various fields: *** Another improved version following the DRY principle. *** Game name: SuperGame Minimum age: 10.5 years and above. Current version: 1.0. It is a AbcCompany product. AbcCompany SuperGame's price details: Version:1.0,cost:$1000 AbcCompany offers a festive season discount. Discounted price detail: Game: SuperGame,Version: 1.0,Discounted price:$800 -----------Changing the game version and price now. Game name: SuperGame Minimum age: 9.5 years and above. Current version: 2.0. It is a AbcCompany product. AbcCompany SuperGame's price details: Version:2.0,cost:$1500 121

Chapter 5

Use the DRY Principle

AbcCompany offers a festive season discount. Discounted price detail: Game: SuperGame,Version: 2.0,Discounted price:$1200 Is this the end of the improvements? You know the answer. There is no end to them; you can always improve your code. Let us think from a general perspective. You know that a company does not finish after making just a single game. It can create multiple games, and can use a common format to display information about them. So, if tomorrow the company wants you to make a new game, say, NewPokemonKid, how should you proceed? Should you copy/ paste the existing code and start editing? You know that this process is not recommended at all. You can make this program better if you move the Game, GameInfo, and GamePrice classes to a shared library and use them accordingly. When you do this, you follow the DRY principle because you do not copy/paste the existing code to make a new game/ new requirement. Instead, you reuse an existing solution that works fine and by using this you indirectly save your test time. So, I create a class library project called BasicGameInfo and then I move these classes into a common file, CommonLibrary.cs (I renamed it from class1.cs). I make these classes public so that I can access them from a different file. For your immediate reference, see the Solution Explorer view in Figure 5-1, where I use a BasicGameInfo project reference in the DryDemoUsingDll project.

Figure 5-1.  DryDemoUsingDll is using the BasicGameInfo project reference 122

Chapter 5

Use the DRY Principle

After I create the project DryDemoUsingDll, I add the BasicGameInfo reference. Figure 5-2 shows a sample snapshot when I right-click the project dependencies, add the reference, and prepare to press the OK button.

Figure 5-2.  Adding BasicGameInfo reference to a C# project file Now I can add using BasicGameInfo; at the beginning of the new file so as to type less. For example, I can directly use Game instead of BasicGameInfo.Game. The same comment applies to GameInfo and GamePrice.

D  emonstration 4 In this sample demo, I change a few parameters, like the name of the games, version, price detail, etc. I put all the pieces together here for your easy reference. Notice that this updated client code is similar to the client code that you saw in the previous demonstration.

123

Chapter 5

Use the DRY Principle

// The content of CommonLibrary.cs using System; namespace BasicGameInfo {     public class Game     {         readonly string companyName;         readonly string gameName;         readonly double minimumAge;         readonly string version;         readonly double actualCost;         readonly double discountedCost;         public Game(                 GameInfo gameInfo,                 GamePrice gamePrice                 )         {             companyName = gameInfo.CompanyName;             gameName = gameInfo.GameName;             version = gameInfo.Version;             minimumAge = gameInfo.MinimumAge;             actualCost = gamePrice.Cost;             discountedCost = gamePrice.DiscountedCost;         }         public void AboutGame()         {             Console.WriteLine($"Game name: {gameName}");             Console.WriteLine($"Minimum age: {minimumAge} years and above.");             Console.WriteLine($"Current version: {version}.");             Console.WriteLine($"It is a {companyName} product.");         }         public void DisplayCost()

124

Chapter 5

Use the DRY Principle

        {             Console.WriteLine($"\n{companyName} {gameName}'s price details:");             Console.WriteLine($"Version:{version},cost:${actualCost}");         }         public void DisplayCostAfterDiscount()         {             Console.WriteLine($"\n{companyName} offers a festive season discount.");             Console.WriteLine("Discounted price detail:");             Console.WriteLine($"Game: {gameName},Version: {version},Discounted price:${discountedCost}");         }     }     public class GameInfo     {         public string CompanyName { get; set; }         public string GameName { get; set; }         public string Version { get; set; }         public double MinimumAge { get; set; }         public GameInfo(string gameName)         {             CompanyName = "AbcCompany";             GameName = gameName;             Version = "1.0";             MinimumAge = 10.5;         }     }     public class GamePrice     {         public double Cost { get; set; }         public double DiscountedCost { get; set; }

125

Chapter 5

Use the DRY Principle

        public GamePrice()         {             Cost = 1000;             DiscountedCost = 800;         }     } } // The content of the new client code using BasicGameInfo; using System; namespace DryDemoUsingDll {     class Program     {         static void Main(string[] args)         {             Console.WriteLine("*** Apply the DRY principle using DLLs. ***");             // Initial setup             GameInfo gameInfo = new GameInfo("NewPokemonKid");             GamePrice gamePrice = new GamePrice();             // Create the game instance with a             // default setup             Game game = new Game(gameInfo,gamePrice);             // Display the default game detail.             game.AboutGame();             game.DisplayCost();             game.DisplayCostAfterDiscount();             Console.WriteLine("------------");             Console.WriteLine("Changing the game version and price now.");

126

Chapter 5

Use the DRY Principle

            // Changing some of the game info             gameInfo.Version = "2.1";             gameInfo.MinimumAge = 12.5;             // Changing the game cost             gamePrice.Cost = 3500;             gamePrice.DiscountedCost = 2000;             // Updating the game instance             game = new Game(gameInfo, gamePrice);             // Display the latest detail             game.AboutGame();             game.DisplayCost();             game.DisplayCostAfterDiscount();             Console.ReadKey();         }     } }

Output When you run this program, you get the following output: *** Apply the DRY principle using DLLs. *** Game name: NewPokemonKid Minimum age: 10.5 years and above. Current version: 1.0. It is a AbcCompany product. AbcCompany NewPokemonKid's price details: Version:1.0,cost:$1000 AbcCompany offers a festive season discount. Discounted price detail: Game: NewPokemonKid,Version: 1.0,Discounted price:$800 ------------

127

Chapter 5

Use the DRY Principle

Changing the game version and price now. Game name: NewPokemonKid Minimum age: 12.5 years and above. Current version: 2.1. It is a AbcCompany product. AbcCompany NewPokemonKid's price details: Version:2.1,cost:$3500 AbcCompany offers a festive season discount. Discounted price detail: Game: NewPokemonKid,Version: 2.1,Discounted price:$2000 You can see that we get the desired output, but this time I used the bare minimum code. I achieved this result by following the DRY principle and reusing existing code. I know what you are thinking. You see a tight coupling between Game and GameInfo/ GamePrice. How can you remove this coupling? Since you learned the DIP in the previous chapter, it should not be a problem for you. I leave this exercise for you.

Summary Code duplication can cause serious problems in software. Programmers often call such duplications evils in the software. Why would we see duplicate code? There are a variety of reasons: some of them are positive and some of them are hard to avoid. By removing redundant code, you make better software that is easier to maintain. This chapter showed you the applications of the DRY principle. You saw that an initial version of a program can be improved multiple times to make it better. Finally, you moved the common code to a shared library. This principle is applicable not only for code, but also for code comments or test cases. For example, you can make a common input file to test various methods instead of passing the same input repeatedly in every method. When you consider using code comments, try to follow the suggestion from the book The Pragmatic Programmer, which tells us to keep low-level knowledge in the code and use the comments for high-level explanations. This perfectly suits the DRY principle’s philosophy; otherwise, for each update, you would need to change both the code and the comments. In short, this principle helps you write cleaner and better code, and thus produce better software. 128

PART III

Make Efficient Applications Part III consists of four chapters, in which we will develop some useful applications that follow several important design patterns. This part will cover the following: •

How can we use factories to separate a code segment that is more likely to vary from a code segment that is less likely to vary?



How can we add new features to an application using wrappers?



How can we use a template method and a hook method together to make an efficient application?



How can we simplify a complex system using facades?

The software industry is full of patterns and design guidelines. As you continue coding and creating different applications, you will discover their importance and understand when to choose one technique over another. In the preface of the book, I told you about the Pareto principle, or 80-20 rule, which states that 80% of outcomes come from 20% of all causes. This is why, in this part, I show you the techniques that are commonly used to build real-world applications. Once you master these techniques, you will get a hint about the answer to the following question: What does a professional programmer think before he or she writes a piece of code?

CHAPTER 6

Separate Changeable Code Using Factories A developer’s end goal is to make an application that meets the customer’s requirements. The code must be easily extensible, maintainable, and stable enough to meet future needs. It is tough to write the best version of a program on the very first attempt. You may need to analyze the code and refactor it several times. In this chapter, you will see such a scenario and learn how to use factories. To make the discussion simple, I start with a small program. We’ll keep analyzing the program and modify it accordingly. To make a stable application, an expert programmer aims to make it loosely coupled. He tries to identify the code that might vary in the future. Once this is done, he separates that part of the code from the remaining codebase. Factories are best in this type of scenario.

POINTS TO REMEMBER The obvious question is: what is a factory? In simple words, this is a code segment that handles the details of an object-creation process. Any method that uses a factory is called a client of the factory. You should note that a factory can have many clients. These clients can use the factory to get an object, and later, as per their needs, they might invoke the methods of that object. So, it is up to the client how to use the object that he receives from a factory. This is why separating the object-creation process is beneficial. Otherwise, you would end up with duplicate code in multiple clients.

© Vaskaran Sarcar 2021 V. Sarcar, Simple and Efficient Programming with C#, https://doi.org/10.1007/978-1-4842-7322-7_6

131

Chapter 6

Separate Changeable Code Using Factories

The Problem Statement Assume that you write a program to demonstrate the behavior of two different animals, say, tigers and cats. But you have a constraint that says that you should not instantiate the animal object inside your Main() method. Why do you see this constraint? Here are some reasons: •

You want to hide the instantiation logic from a client. You know that “change” is the only constant in the programming world. Let’s see what happens if the object’s instantiation logic resides on the client side. When you enhance your application to support a new type of object, you need to update the client code too. It demands retesting the client code as well. How can you separate the instantiation logic from the client code? You’ll see this in the upcoming demonstration.



There may be separate classes with methods that can also create cats or tigers. So, it is better to separate the code that instantiates a cat or a tiger in a commonplace. In such a case, it does not matter how many clients use this code. Every client can refer to the common location to instantiate an object.

I nitial Program I hope you understand the requirement. Let us suppose that you write the following program, shown in Demonstration 1. Before you go through the complete program, note the following points:

132



In this program, the AnimalFactory class is responsible for instantiating an object. It contains a method, called CreateAnimal(), to create a tiger or a cat instance. So, the AnimalFactory acts like a factory class and the CreateAnimal() acts like a factory method.



The CreateAnimal() method is non-static, though you can make it static. I discuss the pros and cons of using a static method in the final chapter (Chapter 11) of this book.

Chapter 6



Separate Changeable Code Using Factories

Inside the client code, you instantiate this factory class to get an animal. This is why, inside the Main() method, you see the following code: AnimalFactory animalFactory = new AnimalFactory(); IAnimal animal = animalFactory.CreateAnimal("cat"); animal.DisplayBehavior();

Demonstration 1 Here is the complete program: using System; namespace UsingSimpleFactory {     interface IAnimal     {         void DisplayBehavior();     }     class Tiger : IAnimal     {         public Tiger()         {             Console.WriteLine("\nA tiger is created.");         }         public void DisplayBehavior()         {             Console.WriteLine("It roars.");             Console.WriteLine("It loves to roam in a jungle.");         }     }     class Cat : IAnimal     {         public Cat()         {             Console.WriteLine("\nA cat is created.");         } 133

Chapter 6

Separate Changeable Code Using Factories

        public void DisplayBehavior()         {             Console.WriteLine("It meows.");             Console.WriteLine("It loves to stay at a home.");         }     }     class AnimalFactory     {         public IAnimal CreateAnimal(string animalType)         {             IAnimal animal;             if (animalType.Equals("cat"))             {                 animal = new Cat();             }             else if (animalType.Equals("tiger"))             {                 animal = new Tiger();             }             else             {                 Console.WriteLine("You can create either a cat or a tiger. ");                 throw new ApplicationException("Unknown animal cannot be instantiated.");             }             return animal;         }     }     class Program     {         static void Main()         {             Console.WriteLine("***Creating animals and learning about them.***");             AnimalFactory animalFactory = new AnimalFactory(); 134

Chapter 6

Separate Changeable Code Using Factories

            IAnimal animal = animalFactory.CreateAnimal("cat");             animal.DisplayBehavior();             animal = animalFactory.CreateAnimal("tiger");             animal.DisplayBehavior();             Console.ReadKey();         }     } }

Output Here is the output: ***Creating animals and learning about them.*** A cat is created. It meows. It loves to stay at a home. A tiger is created. It roars. It loves to roam in a jungle.

Analysis The approach used in Demonstration 1 is quite common in programming. It has a name: simple factory pattern. Now let us analyze this program. You may need to enhance this application if it needs to create a different type of animal, say, monkey, in the future. How would you proceed? You’d need to modify the AnimalFactory class and extend the if-else chain to consider the monkeys. But if you did this, you would violate the OCP, and, as a result, you would need to retest the AnimalFactory class.

135

Chapter 6

Separate Changeable Code Using Factories

POINTS TO REMEMBER When you see switch statements, or an if-else chain to create different types of objects in a similar example, you get a clue that you may need to reopen the code to accommodate future changes. In the worst case, these codes are replicated in several parts of the application. As a result, you keep violating the OCP, which can cause a serious maintenance problem in the future. In this program, Visual Studio shows you a message that says: CA1822: Member 'CreateAnimal' does not access instance data and can be marked as static. It also keeps saying: Active Members that do not access instance data or call instance methods can be marked as static. After you mark the methods as static, the compiler will emit non-­ virtual call sites to these members. This can give you a measurable performance gain for performance-sensitive code. I do not suggest taking this approach now. A static method allows you to call the method without instantiating an object. But a static method has disadvantages too. For example, you cannot change the behavior of a static method at run-time. As said before, you’ll see a detailed discussion about this in Chapter 11.

B  etter Program You understand that by following the OCP principle, you can make the program better. So, in the upcoming demonstration, you will see a new hierarchy. #region Factory hierarchy   abstract class AnimalFactory   {       public abstract IAnimal CreateAnimal();   }   class CatFactory : AnimalFactory   {       public override IAnimal CreateAnimal()       {           return new Cat();       }   } 136

Chapter 6

Separate Changeable Code Using Factories

  class TigerFactory : AnimalFactory   {       public override IAnimal CreateAnimal()       {           return new Tiger();       }   }   #endregion Why is this helpful? I use this construct in such a way that the entire code segment is closed for modification. In the future, if you needed to support a new animal type, say, monkey, you would need to do the following: •

Create a Monkey class that will implement the IAnimal interface.



Create a MonkeyFactory that will implement the AnimalFactory and provide the implementation for the CreateAnimal() method.

So, it is enough for you to test the new classes only. Your existing code is untouched and is closed for modification. Notice that you have two separate inheritance hierarchies in this program: one is for the animal hierarchy and the other is for the factory hierarchy. I have marked them inside this code for your reference. I also include the following class diagram (see Figure 6-1) for better clarity.

Figure 6-1.  The class diagram shows the two different hierarchies in Demonstration 2 137

Chapter 6

Separate Changeable Code Using Factories

Demonstration 2 Here is the complete demonstration: using System; namespace SimpleFactoryModified {     #region Animal hierarchy     interface IAnimal     {         void DisplayBehavior();     }     class Tiger : IAnimal     {         public Tiger()         {             Console.WriteLine("\nA tiger is created.");         }         public void DisplayBehavior()         {             Console.WriteLine("It roars.");             Console.WriteLine("It loves to roam in a jungle.");         }     }     class Cat : IAnimal     {         public Cat()         {             Console.WriteLine("\nA cat is created.");         }         public void DisplayBehavior()         {             Console.WriteLine("It meows.");             Console.WriteLine("It loves to stay at a home.");         } 138

Chapter 6

Separate Changeable Code Using Factories

    }     #endregion     #region Factory hierarchy     abstract class AnimalFactory     {         public abstract IAnimal CreateAnimal();     }     class CatFactory : AnimalFactory     {         public override IAnimal CreateAnimal()         {             return new Cat();         }     }     class TigerFactory : AnimalFactory     {         public override IAnimal CreateAnimal()         {             return new Tiger();         }     }     #endregion     // Client     class Program     {         static void Main()         {             Console.WriteLine("***Modifying the simple factory in the demonstration 1.***");             // The CatFactory creates cats             AnimalFactory animalFactory = new CatFactory();             IAnimal animal = animalFactory.CreateAnimal();             animal.DisplayBehavior();

139

Chapter 6

Separate Changeable Code Using Factories

            // The TigerFactory creates tigers             animalFactory = new TigerFactory();             animal = animalFactory.CreateAnimal();             animal.DisplayBehavior();             Console.ReadKey();         }     } }

Output Except for the first line, this output is the same as the previous output. ***Modifying the simple factory in the demonstration 1.*** A cat is created. It meows. It loves to stay at a home. A tiger is created. It roars. It loves to roam in a jungle.

Analysis You can summarize this modified implementation with the following points:

140



Inside Main(), you decide which animal factory to use—a CatFactory or a TigerFactory?



The subclasses of AnimalFactory create a Cat instance or a Tiger instance.



In this way, you support the concept of the OCP. As a result, you make a better and more extensible solution.

Chapter 6

Separate Changeable Code Using Factories

A New Requirement In Chapter 4, I said: It is not always easy to fully implement this principle, but even partial OCP compliance can be beneficial. Achieving the OCP is not easy for every requirement. A new requirement can demand many changes in an application. In such a case, based on the situation, you must choose one technique over another. For example, let us suppose you have an additional requirement: you want to allow the clients to choose a color for an animal. How should you proceed? One option is to pass the color attribute inside the constructor and update the program accordingly.

D  emonstration 3 Here is a sample demonstration to fulfill the requirement. I have made the important changes in bold. using System; namespace UsingFactoryMethod {     #region Animal hierarchy     interface IAnimal     {         void DisplayBehavior();     }     class Tiger : IAnimal     {         public Tiger(string color)         {             Console.WriteLine($"\nA {color} tiger is created.");         }         public void DisplayBehavior()         {             Console.WriteLine("It roars.");             Console.WriteLine("It loves to roam in a jungle.");         }     } 141

Chapter 6

Separate Changeable Code Using Factories

    class Cat : IAnimal     {         public Cat(string color)         {             Console.WriteLine($"\nA {color} cat is created.");         }         public void DisplayBehavior()         {             Console.WriteLine("It meows.");             Console.WriteLine("It loves to stay at a home.");         }     }     #endregion     #region Factory hierarchy     abstract class AnimalFactory     {         public abstract IAnimal CreateAnimal(string color);     }     class CatFactory : AnimalFactory     {         public override IAnimal CreateAnimal(string color)         {             return new Cat(color);         }     }     class TigerFactory : AnimalFactory     {         public override IAnimal CreateAnimal(string color)         {             return new Tiger(color);         }     }     #endregion

142

Chapter 6

Separate Changeable Code Using Factories

    // Client     class Program     {         static void Main()         {             Console.WriteLine("***Modifying demonstration 2 now.***");             // The CatFactory creates cats             AnimalFactory animalFactory = new CatFactory();             IAnimal animal = animalFactory.CreateAnimal("black");             animal.DisplayBehavior();             // The TigerFactory creates tigers             animalFactory = new TigerFactory();             animal = animalFactory.CreateAnimal("white");             animal.DisplayBehavior();             Console.ReadKey();         }     } }

Output This program produces the following output: ***Modifying demonstration 2 now.*** A black cat is created. It meows. It loves to stay at a home. A white tiger is created. It roars. It loves to roam in a jungle.

143

Chapter 6

Separate Changeable Code Using Factories

Analysis You can see that many changes are required. Is there an alternative way? I think so. Demonstration 4 is made for this purpose. Since the AnimalFactory is an abstract class, you can modify this class to accommodate this change. In this alternative demonstration, I introduce a new method MakeAnimal(), which accepts the color attribute before it calls the CreateAnimal() method to make an animal instance. Here is the code: abstract class AnimalFactory   {       public IAnimal MakeAnimal(string color)       {           Console.WriteLine($"\nThe following animal color is {color}.");           IAnimal animal= CreateAnimal();           return animal;       }       public abstract IAnimal CreateAnimal();   } In the client code, call the MakeAnimal() method instead of CreateAnimal() to see the color of the animal.

Demonstration 4 Here is the modified example. using System; namespace FactoryMethodDemo2 {     #region Animal hierarchy     interface IAnimal     {         void DisplayBehavior();     }     class Tiger : IAnimal     { 144

Chapter 6

Separate Changeable Code Using Factories

        public Tiger()         {             Console.WriteLine("\nA tiger is created.");         }         public void DisplayBehavior()         {             Console.WriteLine("It roars.");             Console.WriteLine("It loves to roam in a jungle.");         }     }     class Cat : IAnimal     {         public Cat()         {             Console.WriteLine("\nA cat is created.");         }         public void DisplayBehavior()         {             Console.WriteLine("It meows.");             Console.WriteLine("It loves to stay at a home.");         }     }     #endregion     #region Factory hierarchy     abstract class AnimalFactory     {         public IAnimal MakeAnimal(string color)         {             Console.WriteLine($"\nThe following animal color is {color}.");             IAnimal animal= CreateAnimal();             return animal;         }         public abstract IAnimal CreateAnimal();     } 145

Chapter 6

Separate Changeable Code Using Factories

    class CatFactory : AnimalFactory     {         public override IAnimal CreateAnimal()         {             return new Cat();         }     }     class TigerFactory : AnimalFactory     {         public override IAnimal CreateAnimal()         {             return new Tiger();         }     }     #endregion     // Client     class Program     {         static void Main()         {             Console.WriteLine("***Modifying demonstration 2 now.***");             // The CatFactory creates cats             AnimalFactory animalFactory = new CatFactory();             IAnimal animal = animalFactory.MakeAnimal("black");             animal.DisplayBehavior();             // The TigerFactory creates tigers             animalFactory = new TigerFactory();             animal = animalFactory.MakeAnimal("white");             animal.DisplayBehavior();             Console.ReadKey();         }     } }

146

Chapter 6

Separate Changeable Code Using Factories

Output Here is the output: ***Modifying demonstration 2 now.*** The following animal color is black. A cat is created. It meows. It loves to stay at a home. The following animal color is white. A tiger is created. It roars. It loves to roam in a jungle.

Analysis At the beginning of this chapter, we looked at the advantages of using factories. Starting from Demonstration 2, we used a new hierarchy for factories such that where all concrete factories inherit from AnimalFactory, we pass the details of object creation to a concrete factory (CatFactory or TigerFactory). Since we are following the OCP principle, we could add a new concrete factory, say, MonkeyFactory, to create monkeys. If we implemented this scenario, we would not need to reopen existing code. Instead, the new MonkeyFactory class could inherit from AnimalFactory, and, following the rules, it would create monkeys. In this case, we would need to create a Monkey class in the same way we made the Tiger class or the Cat class. Notice that we would never need to reopen the existing code. I created Demonstrations 3 and 4 to support a new requirement. Maintaining the OCP in the current structure to accommodate the new requirement was tough because the color attribute was not considered at the beginning. Demonstration 4 shows you that you can still proceed with minimum changes.

147

Chapter 6

Separate Changeable Code Using Factories

Summary Factories provide an alternative method for object creation. This chapter starts with a simple factory class. It helps you to separate the code that is likely to vary from the other parts of the code. You put the instantiation logic inside the factory class to provide a uniform way for object creation. Following the OCP principle, you further modified the application. You created another hierarchy for factories and passed the actual object-creation logic to the concrete factories. Later, in Demonstration 4, you saw that inside the abstract factory class, you can set a common rule that all derived concrete factories must follow. This process can help you accommodate particular requirement(s) with minimum changes.

148

CHAPTER 7

Add Features Using Wrappers An alternative to inheritance is composition. It is quite common in programming and often gives you a better payoff. This chapter shows you a useful case study on this topic using some wrappers. The first question that may come into your mind is: what is a wrapper? A wrapper is like a topping that surrounds an object. In programming, you often use a wrapper to add some functionalities dynamically. This is a powerful technique because you can add or discard a wrapper as per your needs, and it does not hamper the functionalities of the original object. Consider a case: You need to work on a piece of code and add some new features. Someone coded this earlier, and you cannot change the existing code. This scenario is common in the software industry when you need to enhance a feature to attract new customers, but you cannot alter the existing workflow of the software, because you must still support existing customers. In this case, since you were not a part of the team that wrote the first version of the software, you did not have exclusive control from the beginning. Wrappers are useful in such situations. As said before, in this case, you can add new functionality on top of the existing functionality to support new customers. In fact, using different types of wrappers, you can add different types of customers too. The upcoming example will make the concept clearer to you.

© Vaskaran Sarcar 2021 V. Sarcar, Simple and Efficient Programming with C#, https://doi.org/10.1007/978-1-4842-7322-7_7

149

Chapter 7

Add Features Using Wrappers

The Problem Statement Consider a group of people who each want to purchase property and have their own home. The budget and mindset of each individual are different. So, this group of people visits a homebuilder to get a cost estimate. For simplicity, let us assume that they have the following options: •

They can either make a basic home with minimum facilities or they can make an advanced home with more facilities. For illustration purposes, let us refer to these homes as the BasicHome and AdvancedHome, respectively.



The homebuilder gives them options: a customer can opt for a playground, or a swimming pool, or both. Let us call these luxuries. Each of these luxuries adds additional costs for a buyer.

Based on budget restrictions, a customer can opt for various options, and the final price will vary. Most important, a customer who opts for a BasicHome today can upgrade his home tomorrow by adding a playground or a swimming pool (or both). Can you write a program for this scenario?

Using Subclassing If you try to provide a solution using inheritance, you’ll discover the problems that are associated with it. Assume that you start with the following structure: class Home {     // Some code } class PlayGround : Home {     // Some code } class SwimmingPool : PlayGround {     // Some code } 150

Chapter 7

Add Features Using Wrappers

This is not a recommended approach, because to get a swimming pool, you first must get a playground, which a customer may not want. Due to similar logic, the following structure is not a good choice either: class Home {     // Some code } Class SwimmingPool : Home {     // Some code } class PlayGround : SwimmingPool {     // Some code } This is because in this case to get a playground you first must get a swimming pool, which a customer may not want. So, implementing a multilevel inheritance, in this case, is not a good idea! Now, let’s assume that you start with a hierarchical inheritance where SwimmingPool and PlayGround both inherit from the Home class, as shown in Figure 7-1.

Figure 7-1.  A hierarchical inheritance Now you would get a home with a swimming pool and a playground. So, you would end up with the design shown in Figure 7-2.

151

Chapter 7

Add Features Using Wrappers

Figure 7-2.  A class needs to inherit from multiple base classes. It causes the diamond problem in C# But you know that you cannot have multiple base classes in C#. So, any construct like the following will raise a compilation error too: class Home: SwimmingPool, PlayGround // Error { } You can see that using simple subclassing, in this case, is not a good idea. What are the alternatives? Let us continue investigating. You may proceed with an interface for the luxuries. For example, you can opt for the following interface: interface ILuxury {     void AddPlayGround();     void AddSwimmingPool(); }

152

Chapter 7

Add Features Using Wrappers

Now you need a class that can implement this interface. For example, here is a Home class that extends the BasicHome class and implements the ILuxury interface: class Home : BasicHome, ILuxury {     public void AddPlayGround()     {         // Some code     }     public void AddSwimmingPool()     {         // Some code     } } But again, a customer may opt for a home with one of these luxuries, but not both. In that case, if a method is not needed, you would write: throw new NotImplementedException(); The problem associated with this is discussed in the context of LSP in Chapter 4. To avoid this, you may follow ISP and segregate the ILuxury interface. Yes, this time it can work! Since you saw a similar solution (when I considered different capabilities into a separate hierarchy) in Chapter 2, I’ll not repeat it here. Next, we look for an alternative approach. This chapter is made for this.

Using Object Composition Let us see how a wrapper can help you. Using a wrapper, you surround an object with another object. The enclosing object is often called a decorator and conforms to the interface of the component that it decorates. It forwards the requests to the original component and can perform additional operations before or after those requests. You can add an unlimited number of responsibilities with this concept. The following figures help you understand this. Figure 7-3 shows the home (basic or advanced) is surrounded by a playground.

153

Chapter 7

Add Features Using Wrappers

Figure 7-3.  The home is surrounded by a playground Figure 7-4 shows the home is surrounded by a swimming pool.

Figure 7-4.  The home is surrounded by a swimming pool Figure 7-5 shows the home is surrounded by a playground and a swimming pool. Here you first surround the home with a playground and then you surround the structure with a swimming pool.

154

Chapter 7

Add Features Using Wrappers

Figure 7-5.  The home is surrounded by a playground and a swimming pool Figure 7-6 shows the home is surrounded by a swimming pool and a playground again. But this time you change the order; you first surround the home with a swimming pool, and then you surround it with a playground.

Figure 7-6.  The home is surrounded by a swimming pool. Later you surround the structure with a playground

Note  Following the same technique, you can add two more playgrounds or swimming pools.

155

Chapter 7

Add Features Using Wrappers

Let us try to implement this concept following the requirement we have. In the upcoming demonstration, six players are involved: Home, BasicHome, AdvancedHome, Luxury, PlayGround, SwimmingPool. Home is defined as follows: abstract class Home     {         public double basePrice = 100000;         public double AdditionalCost { get; set; }         public abstract void BuildHome();         public virtual double GetPrice()         {             return basePrice + AdditionalCost;         }     } Here are some important points:

156



You can see that a concrete implementor of Home must implement the BuildHome() and GetPrice() methods. In this example, BasicHome and AdvancedHome inherit from Home.



I assume that the base price of a home is $100,000. Using the AdditionalPrice property, one can set some extra prices. I use this property to set an additional cost for an advanced home. Currently, for a basic home, this cost is 0, and for an advanced home, this cost is $25,000.



I assume that once the home is built there is no need for an immediate modification. One can add the luxuries later.



Once a home is built, you can opt for a playground or a swimming pool for an existing home, or you may want both. So, the PlayGround and SwimmingPool classes appear in this example.



Though it was not strictly needed, to share common code, both the PlayGround class and SwimmingPool class inherit from the abstract class Luxury, which has the following structure:

Chapter 7

Add Features Using Wrappers

   abstract class Luxury : Home     {         protected Home home;         public double LuxuryCost { get; set; }         public Luxury(Home home)         {             this.home = home;         }         public override void BuildHome()         {             home.BuildHome();         }     } •

Like the AdditionalPrice property, one can set/update a luxury cost using the LuxuryCost property.



Notice that Luxury holds a reference of Home. So, the concrete decorators (PlayGround or SwimmingPool in this example) are decorating an instance of Home.



Now let’s look into the structure of a concrete decorator, say, PlayGround, which is as follows: class PlayGround : Luxury     {         public PlayGround(Home home) : base(home)         {             this.LuxuryCost = 20000;         }         public override void BuildHome()         {             base.BuildHome();             AddPlayGround();         }

157

Chapter 7

Add Features Using Wrappers

        private void AddPlayGround()         {             Console.WriteLine($"For a playground,you pay extra ${ this.LuxuryCost}.");             Console.WriteLine($"Now the total cost is: ${GetPrice()}.");         }         public override double GetPrice()         {             return home.GetPrice() + LuxuryCost;         }     } •

You can see that by using the AddPlayGround() method, you can add a playground. When you add this facility, you have to pay an additional $20,000. I initialize this value inside the constructor. Most important, notice that before adding a playground, it calls BuildHome() from the base class Luxury. This method in turn calls BuildHome() from a concrete implementation of Home.



The SwimmingPool class works similarly, but you have to pay more for this. (Yes, I assume that a swimming pool costs more than a playground in this case).

C  lass Diagram Figure 7-7 shows the most important parts of the class diagram.

158

Chapter 7

Add Features Using Wrappers

Figure 7-7.  The class diagram shows the participants except for the client class

Demonstration Here is the complete demonstration for you. In the client code, you can see many different scenarios to show the effectiveness of this application. using System; namespace UsingWrappers {     abstract class Home     {         public double basePrice = 100000;         public double AdditionalCost { get; set; }         public abstract void BuildHome();         public virtual double GetPrice()         {             return basePrice + AdditionalCost;         }     } 159

Chapter 7

Add Features Using Wrappers

    class BasicHome : Home     {         public BasicHome()         {             AdditionalCost = 0;         }         public override void BuildHome()         {             Console.WriteLine("A home with basic facilities is made.");             Console.WriteLine($"It costs ${GetPrice()}.");         }     }     class AdvancedHome : Home     {         public AdvancedHome()         {             AdditionalCost = 25000;         }         public override void BuildHome()         {             Console.WriteLine("A home with advanced facilities is made.");             Console.WriteLine($"It costs ${GetPrice()}.");         }     }     abstract class Luxury : Home     {         protected Home home;         public double LuxuryCost { get; set; }         public Luxury(Home home)         {             this.home = home;         }

160

Chapter 7

Add Features Using Wrappers

        public override void BuildHome()         {             home.BuildHome();         }     }     class PlayGround : Luxury     {         public PlayGround(Home home) : base(home)         {             this.LuxuryCost = 20000;         }         public override void BuildHome()         {             base.BuildHome();             AddPlayGround();         }         private void AddPlayGround()         {             Console.WriteLine($"For a playground, you pay an extra ${ this.LuxuryCost}.");             Console.WriteLine($"Now the total cost is ${GetPrice()}.");         }         public override double GetPrice()         {             return home.GetPrice() + LuxuryCost;         }     }     class SwimmingPool : Luxury     {         public SwimmingPool(Home home) : base(home)         {             this.LuxuryCost = 55000;         }

161

Chapter 7

Add Features Using Wrappers

        public override void BuildHome()         {             base.BuildHome();             AddSwimmingPool();         }         private void AddSwimmingPool()         {             Console.WriteLine($"For a swimming pool, you pay an extra ${this.LuxuryCost}.");             Console.WriteLine($"Now the total cost is ${GetPrice()}.");         }         public override double GetPrice()         {             return home.GetPrice() + LuxuryCost;         }     }     class Program     {         static void Main(string[] args)         {             Console.WriteLine("***Using wrappers.***");             Console.WriteLine("Scenario-1: A basic home with basic facilities.");             Home home = new BasicHome();             home.BuildHome();             Console.WriteLine("\nScenario-2: A basic home with an additional playground.");             Luxury homeWithOnePlayGround = new PlayGround(home);             homeWithOnePlayGround.BuildHome();             Console.WriteLine("\nScenario-3: A basic home with two additional playgrounds.");             Luxury homeWithDoublePlayGrounds = new PlayGround(homeWithOne PlayGround);             homeWithDoublePlayGrounds.BuildHome(); 162

Chapter 7

Add Features Using Wrappers

            Console.WriteLine("\nScenario-4: A basic home with one additional playground and swimming pool.");             Luxury homeWithOnePlayGroundAndOneSwimmingPool = new Swimming Pool(homeWithOnePlayGround);             homeWithOnePlayGroundAndOneSwimmingPool.BuildHome();             Console.WriteLine("\nScenario-5: Adding a swimming pool and then a playground to a basic home.");             Luxury homeWithOneSimmingPool = new SwimmingPool(home);             Luxury homeWithSwimmingPoolAndPlayground = new PlayGround( homeWithOneSimmingPool);             homeWithSwimmingPoolAndPlayground.BuildHome();             Console.WriteLine("\nScenario-6: An advanced home with some more facilities.");             home = new AdvancedHome();             home.BuildHome();             Console.WriteLine("\nScenario-7: An advanced home with an additional playground.");             homeWithOnePlayGround = new PlayGround(home);             homeWithOnePlayGround.BuildHome();             Console.ReadKey();         }     } }

Output Here is the output: ***Using wrappers.*** Scenario-1: A basic home with basic facilities. A home with basic facilities is made. It costs $100000.

163

Chapter 7

Add Features Using Wrappers

Scenario-2: A basic home with an additional playground. A home with basic facilities is made. It costs $100000. For a playground, you pay an extra $20000. Now the total cost is $120000. Scenario-3: A basic home with two additional playgrounds. A home with basic facilities is made. It costs $100000. For a playground, you pay an extra $20000. Now the total cost is $120000. For a playground, you pay an extra $20000. Now the total cost is $140000. Scenario-4: A basic home with one additional playground and swimming pool. A home with basic facilities is made. It costs $100000. For a playground, you pay an extra $20000. Now the total cost is $120000. For a swimming pool, you pay an extra $55000. Now the total cost is $175000. Scenario-5: Adding a swimming pool and then a playground to the basic home. A home with basic facilities is made. It costs $100000. For a swimming pool, you pay an extra $55000. Now the total cost is $155000. For a playground, you pay an extra $20000. Now the total cost is $175000. Scenario-6: An advanced home with some more facilities. A home with advanced facilities is made. It costs $125000.

164

Chapter 7

Add Features Using Wrappers

Scenario-7: An advanced home with an additional playground. A home with advanced facilities is made. It costs $125000. For a playground, you pay an extra $20000. Now the total cost is $145000.

A  nalysis Notice that this implementation follows the OCP principle. So, when you make a different type of home, you do not need to open the existing code, but rather you can make a new class that inherits from the abstract class Home. I want you to note that I slightly violate SRP in this example. This is because I wanted to show you the final price after uniformly adding a luxury. In actuality, when I add a wrapper, I do not need to calculate the total cost; instead, it is sufficient to show the increased price for this addition. But I assume that a customer will like to see the total estimated price. This is the reason, after each addition of a luxury item, I showed the total cost. In addition, the price depends on the type of home you choose. So, it makes sense to put the GetPrice() method and BuildHome() method in the Home class. This example pattern is more effective when you use a wrapper that does a single job. So, when you follow the SRP, you can add or remove a behavior easily using this kind of wrapper. I am about to finish this chapter. But first, I want to inform you that when you analyze I/O stream implementations in .NET Framework and Java, you will find many similar implementations. For example, the BufferedStream class inherits from the abstract base class Stream. I am taking a snapshot from Visual Studio IDE to show you the constructors of this class (see Figure 7-8).

Figure 7-8.  Partial snapshot of BufferedStream class from Visual Studio IDE

165

Chapter 7

Add Features Using Wrappers

You can see that the BufferedStream class constructors can accept a Stream class object as a parameter. Notice that the Luxury constructor also accepts its base class object (Home) as a parameter. So, you get a clue that BufferedStream class follows the wrapper pattern. But now notice the partial snapshot of the FileStream class from Visual Studio IDE (see Figure 7-9).

Figure 7-9.  Partial snapshot of FileStream class from Visual Studio IDE You can see that there is no constructor of FileStream class that can accept a Stream class object as a parameter. So, this class does not follow the wrapper/decorator pattern.

S  ummary The pattern shown in this chapter is referred to as a wrapper or a decorator pattern. This chapter shows you an alternative to the subclassing technique. You have seen the reason why neither multilevel inheritance nor multiple inheritance can solve the problem that we mentioned at the beginning of the chapter. Later, you saw an implementation using object composition. You used different types of wrappers to add a behavior dynamically in this application. Recall that a simple inheritance promotes only a compile-time binding, but not a dynamic binding.

166

Chapter 7

Add Features Using Wrappers

In short, here are the key points that you learned in this chapter: •

You can add a state and/or behavior without modifying an existing inheritance hierarchy.



In Demonstration 1, you defined a new hierarchy that itself extends the root of the original/existing hierarchy.



To use the decorators, you first instantiated a home and then wrapped it inside a decorator.



This example pattern has a name. We call it the decorator pattern. It shows an example of when object composition can perform better than plain inheritance.

You have also seen some in-built examples in .NET. There you learned that the BufferedStream class follows a similar pattern, but the FileStream class does not.

167

CHAPTER 8

Efficient Templates Using Hooks This chapter will show you two important techniques. First, you’ll learn to use a template method. Why is this important? Template methods are one of the fundamental techniques for code reuse. Suppose you follow a multi-step algorithm to achieve a task. Using a template method, you can redefine some of these steps (but not all them) without altering their calling sequence. This chapter starts with a demonstration that uses a template method. Later, you’ll use a hook method in addition to the template method to enhance this application.

The Problem Statement A dealer (or seller) sells various products such as televisions, washing machines, and music players. Assume that you know such a dealer. You can visit his showroom to purchase a television. You can visit the same showroom to purchase a washing machine for your home. In each case, you can summarize the overall activity in the following order: 1: 2: 3: 4:

You You The The

visit the dealer showroom. purchase a product. dealer generates a bill(or,invoice) for you. dealer delivers the product to you.

Can you make an application that mimics this scenario? POINT TO NOTE  Many of us differentiate a bill from an invoice. I suggest you not to put focus on the difference. Please treat them same in this chapter.

© Vaskaran Sarcar 2021 V. Sarcar, Simple and Efficient Programming with C#, https://doi.org/10.1007/978-1-4842-7322-7_8

169

Chapter 8

Efficient Templates Using Hooks

Initial Program In the upcoming program, I assume that you purchase a washing machine and television from the same dealer. When you purchase a television, you see the following output: 1.The 2.The 3.The 4.The

customer visits a dealer showroom. customer purchases a television. bill is printed. product is delivered.

When you purchase a washing machine, you see the following output: 1.The 2.The 3.The 4.The

customer visits a dealer showroom. customer purchases a washing machine. bill is printed. product is delivered.

Notice that steps 1, 3, and 4 are common to both cases. You do not see any difference in these steps when you purchase different products. Also notice that these steps are fixed. For example, once you purchase a product, the bill is generated and then the product is delivered. The system is unlikely to generate a bill and deliver the product if you do not first visit the showroom or choose the product. (I do not consider online shopping in this case). A template method is ideal in this scenario, where you do not alter the basic steps of an algorithm but do allow some minor modifications in some steps. In our example, step 2 changes slightly to show the product you select, but the remaining steps are the same when you buy any product. When you order a pizza, you notice a similar scenario. For example, you can opt for different toppings, such as bacon, onions, extra cheese, or mushrooms, for the pizza. How does the chef make the pizza? He first prepares the pizza following his conventional ways. Just before delivery, he adds the toppings to make you happy. You can find similar situations in other domains too. Now the question is: how can you make an application that has multiple steps, but only a few of them vary? The answer is: you can use a template method (which consists of the many steps) in a parent class. Then you can defer some steps to the subclasses (which represent the particular products) and allow them to override those steps as needed. 170

Chapter 8

Efficient Templates Using Hooks

Note  Using simple polymorphism, you can bring about a radical change by overriding all or most of the methods of a parent class inside a child class. However, when you use a template method, you do not override all the parent class (or base class) methods in the child class. Instead, you override only a limited number of methods (or steps). This is the key distinction between this approach and simple polymorphism. In the upcoming example, you can see the following parent class that contains a template method called PurchaseProduct. I have used comments for your easy understanding. public abstract class Device     {         // The following method(step) will NOT vary         private void VisitShowroom()         {             Console.WriteLine("1.The customer visits a dealer showroom.");         }         // The following method(step) will NOT vary         private void GenerateBill()         {             Console.WriteLine("3.The bill is printed.");         }         private void DeliverProduct()         {             Console.WriteLine("4.The product is delivered.\n");         }         /*         The following method will vary. It will be         overridden by derived classes.         */         protected abstract void SelectProduct();         // The template method 171

Chapter 8

Efficient Templates Using Hooks

        public void PurchaseProduct()         {             // Step-1             VisitShowroom();             // Step-2: Specialized action             SelectProduct();             // Step-3             GenerateBill();             // Step-4             DeliverProduct();         }     } Here are the important points:

172



This parent class is abstract because it contains the abstract method SelectProduct(). A derived class overrides this method to show the product you purchase.



Notice that inside the template method there are four methods: VisitShowroom(), SelectProduct(), GenerateBill(), DeliverProduct(). These four methods represent the four steps of the algorithm.



SelectProduct() is a protected method. It allows a derived class to redefine/override the method. But the other methods inside the template method are marked with the private keyword/access modifier. So, a client cannot access them directly.



When you call the PurchaseProduct()method, a derived class cannot alter the execution order of these methods. Also, you cannot access these methods directly inside the client code (i.e., inside the Main() method). To complete a purchase, you need to invoke the template method. This is why I made this template method public. From a client’s perspective, he does not know how the template method accomplishes its task, since you do not expose the inner logic to the client. This is a better practice.

Chapter 8

Efficient Templates Using Hooks

C  lass Diagram The following figure (Figure 8-1) shows the important parts of the class diagram.

Figure 8-1.  PurchaseProduct() is the template method in this example

D  emonstration 1 Here is the complete demonstration: using System; namespace TemplateMethodDemo {     ///     /// Basic skeleton of action steps     ///     public abstract class Device     {

173

Chapter 8

Efficient Templates Using Hooks

        // The following method(step) will NOT vary         private void VisitShowroom()         {             Console.WriteLine("1.The customer visits a dealer showroom.");         }         // The following method(step) will NOT vary         private void GenerateBill()         {             Console.WriteLine("3.The bill is printed.");         }         private void DeliverProduct()         {             Console.WriteLine("4.The product is delivered.\n");         }         /*         The following method will vary. It will be         overridden by derived classes.         */         protected abstract void SelectProduct();         // The template method         public void PurchaseProduct()         {             // Step-1             VisitShowroom();             // Step-2: Specialized action             SelectProduct();             // Step-3             GenerateBill();             // Step-4             DeliverProduct();         }     }

174

Chapter 8

Efficient Templates Using Hooks

    // The concrete derived class-Television     public class Television : Device     {         protected override void SelectProduct()         {             Console.WriteLine("2.The customer purchases a television.");         }     }     // The concrete derived class-WashingMachine     public class WashingMachine : Device     {         protected override void SelectProduct()         {              Console.WriteLine("2.The customer purchases a washing machine.");         }     }     class Program     {         static void Main(string[] args)         {             Console.WriteLine("***A demonstration of a template Method.***\n");             Console.WriteLine("---The customer wants a television.---");             Device device = new Television();             device.PurchaseProduct();              Console.WriteLine("---The customer wants a washing machine.---");             device = new WashingMachine();             device.PurchaseProduct();             Console.ReadLine();         }     } }

175

Chapter 8

Efficient Templates Using Hooks

Output Here is the output: ***A demonstration of a template Method.*** ---The customer wants a television.--1.The customer visits a dealer showroom. 2.The customer purchases a television. 3.The bill is printed. 4.The product is delivered. ---The customer wants a washing machine.--1.The customer visits a dealer showroom. 2.The customer purchases a washing machine. 3.The bill is printed. 4.The product is delivered.

Analysis You can see that, in the future, if you need to consider a different product, say SmartPhone, you can easily enhance the application. In that case, you can create a SmartPhone class that inherits from Device and overrides SelectProduct() in the same way. Here is a sample for you: // The concrete derived class-SmartPhone public class SmartPhone : Device {     protected override void SelectProduct()     {         Console.WriteLine("2.The customer purchases a smartphone.");     } }

176

Chapter 8

Efficient Templates Using Hooks

Note This implementation obeys the OCP principle. But does this implementation violate the SRP? The answer seems to be yes to some extent. But think from a seller’s perspective: a potential customer visits the showroom and selects the product. Then the seller generates the invoice and delivers the product to the customer. All these activities are linked to “one successful sale” from the seller’s point of view. From a programmer’s perspective, all these steps are combined to accomplish one single task: purchase a product. This is why a customer in this example can access only the template method, and other methods are hidden to him. In addition, you may recall what I said in the preface: sometimes it is OK to bend a rule based on the complexity or the nature of a problem.

Enhanced Requirement Let us enhance the application for another real-world scenario. The dealer can decide to offer a special coupon to any customer who purchases a television from him. This offer does not apply to other products. How can you modify this application to serve this new requirement? One approach is straightforward. You can use a method (it can be public or protected) to reflect this offer and place the method inside the parent class Device. Let’s name this method GenerateGiftCoupon(). Here is a sample code for this: /// /// Basic skeleton of action steps /// public abstract class Device {     // The following method(step) will NOT vary     private void VisitShowroom()     {     Console.WriteLine("1.The customer visits a dealer showroom.");     }     // The following method(step) will NOT vary

177

Chapter 8

Efficient Templates Using Hooks

    private void GenerateBill()     {     Console.WriteLine("3.The bill is printed.");     }     private void DeliverProduct()     {     Console.WriteLine("4.The product is delivered.\n");     }     /*     The following method will vary. It will be     overridden by derived classes.     */     protected abstract void SelectProduct();     // The template method     public void PurchaseProduct()     {         // Step-1         VisitShowroom();         // Step-2: Specialized action         SelectProduct();         // Step-2.1: Elgible for a gift?         GenerateGiftCoupon();         // Step-3         GenerateBill();         // Step-4         DeliverProduct();     }     protected virtual void GenerateGiftCoupon()     {       Console.WriteLine("A gift coupon is generated.");     } }

178

Chapter 8

Efficient Templates Using Hooks

Now any subclass of Device can have the GenerateGiftCoupon() method. They can redefine it as per their own needs. So now, as per our new requirement, you get a special coupon from the dealer if you purchase a television, but not for buying a washing machine. So, inside the WashingMachine class, you override the method and write it in the following way: protected override void GenerateGiftCoupon() {   throw new NotImplementedException(); } But throwing an exception inside a method body can be risky in certain scenarios. You learned about this when I discussed the Liskov substitution principle (LSP) in Chapter 4. To avoid this problem, you can make this method empty as follows: protected override void GenerateGiftCoupon() {   // Empty body } Now think: is it a good idea to use an empty method? I do not think so. Let us look at an alternative. You may want to make GenerateGiftCoupon() abstract and override it in its child classes as needed. Yes, this will work. But the problem is that when you use an abstract method in the parent class, the derived classes need to provide the concrete implementation for the method (otherwise, it is again abstract, and you cannot instantiate from it). So, if you have too many specialized classes and most of them don’t make you eligible for gift coupons, you are still forced to override them. (Can you remember ISP?) Is there a better solution? Yes, I think so. You can use a hook method. I am about to show this in Demonstration 2. But what is a hook in programming? In very simple words, a hook helps you execute some code before or after existing code. It can help you extend the behavior of a program at run-time. Hook methods can provide some default behaviors that a subclass can override if necessary. Often, they do nothing by default.

179

Chapter 8

Efficient Templates Using Hooks

Let me show you the use of a simple hook in this program. Notice the bold lines in the following code segments: // The template method public void PurchaseProduct() {     // Step-1     VisitShowroom();     //Step-2: Specialized action     SelectProduct();     // Step-2.1: Elgible for a gift?     if(IsEligibleForGiftCoupon())     {         GenerateGiftCoupon();     }     // Step-3     GenerateBill();     // Step-4     DeliverProduct(); } Where the hook method is defined as: // If a customer purchases a television // he can get a gift. By default, // there is no gift coupon. protected virtual bool IsEligibleForGiftCoupon() {   return false; } These two code segments tell us that when you invoke the template method, by default GenerateGiftCoupon() will not be executed. This is because IsEligibleForGiftCoupon() returns false, which in turn makes the if condition inside the template method false. But the Television class overrides this method as follows:

180

Chapter 8

Efficient Templates Using Hooks

protected override bool IsEligibleForGiftCoupon() {   return true; } So, when you instantiate a Television object and call the template method, GenerateGiftCoupon() is called just before step 3.

Demonstration 2 Here is the complete demonstration using a hook method. I keep the comments for your easy understanding. using System; namespace UsingHook {     ///     /// Basic skeleton of action steps     ///     public abstract class Device     {         // The following method (step) will NOT vary         private void VisitShowroom()         {             Console.WriteLine("1.The customer visits a dealer showroom.");         }         // The following method (step) will NOT vary         private void GenerateBill()         {             Console.WriteLine("3.The bill is printed.");         }         private void DeliverProduct()         {             Console.WriteLine("4.The product is delivered.\n");         }

181

Chapter 8

Efficient Templates Using Hooks

        /*         The following method will vary. It will be         overridden by derived classes.         */         protected abstract void SelectProduct();         // The template method         public void PurchaseProduct()         {             // Step-1             VisitShowroom();             // Step-2: Specialized action             SelectProduct();             // Step-2.1: Elgible for gift?             if(IsEligibleForGiftCoupon())             {                 GenerateGiftCoupon();             }             // Step-3             GenerateBill();             // Step-4             DeliverProduct();         }         protected void GenerateGiftCoupon()         {             Console.WriteLine("A gift coupon is generated.");         }         //         //         //         //

182

Hook If a customer purchases a television he can get a gift. By default, there is no gift coupon.

Chapter 8

Efficient Templates Using Hooks

        protected virtual bool IsEligibleForGiftCoupon()         {             return false;         }     }     // The concrete derived class-Television     public class Television : Device     {         protected override bool IsEligibleForGiftCoupon()         {             return true;         }         protected override void SelectProduct()         {             Console.WriteLine("2.The customer purchases a television.");         }     }     // The concrete derived class-WashingMachine     public class WashingMachine : Device     {         protected override void SelectProduct()         {             Console.WriteLine("2.The customer purchases a washing machine.");         }     }     class Program     {         static void Main(string[] args)         {             Console.WriteLine("***A demonstration of a template Method.***\n");             Console.WriteLine("---The customer wants a television.---"); 183

Chapter 8

Efficient Templates Using Hooks

            Device device = new Television();             device.PurchaseProduct();              Console.WriteLine("---The customer wants a washing machine.---");             device = new WashingMachine();             device.PurchaseProduct();             Console.ReadLine();         }     } }

Output Here is the output: ***A demonstration of a template Method.*** ---The customer wants a television.--1.The customer visits a dealer showroom. 2.The customer purchases a television.   A gift coupon is generated. 3.The bill is printed. 4.The product is delivered. ---The customer wants a washing machine.--1.The customer visits a dealer showroom. 2.The customer purchases a washing machine. 3.The bill is printed. 4.The product is delivered.

Summary This chapter shows you how to use a template method to make an efficient application. Later, it demonstrates how to use hooks to adopt a new requirement without altering the core structure of an algorithm.

184

Chapter 8

Efficient Templates Using Hooks

Wikipedia says that Microsoft Windows also allows you to insert hooks. You can use them to insert, remove, process, or modify keyboard and mouse events. There is a downside though. If you are not careful enough, the use of hooks can impact the overall performance of your application. But in our case, using a hook method was beneficial. In a similar situation, using a hook method, you can extend the application to accommodate a new requirement.

185

CHAPTER 9

Simplify Complex Systems Using Facades In Chapter 8, you saw how you can redefine some steps in a multi-step algorithm to achieve a task. In this chapter, you will see an application that also does a series of tasks. But instead of redefining some of these tasks, you will make a simplified interface to perform them. Facades are useful in this context.

Note  You may be interested to learn the difference between a facade and a template method. In general, a template method belongs to a base class and allows the subclasses to redefine some steps. You create an object of a class and invoke this template method to complete your job. But facades often involve multiple objects from many different classes. This time, you perform a series of steps to accomplish the task involving all these objects. You do not redefine the methods in these classes; instead, you manage to call them easily. So, a subsystem class often does not know about the presence of a facade. Facades provide you with an entry point to access various methods across different classes in a structured manner. If you enforce a rule that does not allow you access to the individual methods directly, and instead you access them through only your facade, then the facade is called an opaque facade; otherwise, it is a transparent facade. You can make your facade static too.

© Vaskaran Sarcar 2021 V. Sarcar, Simple and Efficient Programming with C#, https://doi.org/10.1007/978-1-4842-7322-7_9

187

Chapter 9

Simplify Complex Systems Using Facades

The Problem Statement A person can apply for a loan from a bank. The bank must do some background verification before it grants the loan to the customer. This background verification is a complex process that consists of various subprocesses. Bank officials can visit a customer’s property too before they consider the loan application. If an applicant fulfills all these criteria, he can get a loan. But here is the key: an applicant does not know the details of the background verification, and he is interested only in the final outcome— whether he can get the loan. How the bank officials reach a decision does not matter to the customer. In the upcoming examples, you will see such a process. For simplicity, I make the following assumptions: •

A loan applicant or customer must have some assets. If the asset value is less than the loan amount he seeks, he cannot get the loan.



If a customer has an existing loan, he cannot get any new loan.

Our job is to make an application that is based on these assumptions. Here is a sample output for better clarity: Case-1: Bob’s current asset value: USD 5000 He claims loan amount: USD 20000 He has an existing loan. Expected Outcome: Bob cannot get the loan for the following reasons: •

Insufficient balance.



Old loan exists.

Case-2: Jack's current asset value: USD 100000 He claims loan amount: USD 30000 He has no existing loan.

188

Chapter 9

Simplify Complex Systems Using Facades

Expected Outcome: Jack can get the loan. Case-3: Tony's current asset value: USD 125000 He claims loan amount: USD 50000 He has an existing loan. Expected Outcome: Tony cannot get the loan for the following reason: •

Old loan exists.

Let us build the application.

Initial Program In this example, you see three classes: Person, Asset, and LoanStatus. An instance of the Person class can apply for a loan. This class has a constructor that takes three parameters: name, assetValue, and previousLoanExist. To avoid more typing and passing all three parameters during the instance creation, I made the last two parameters optional. Here is the Person class: class Person     {         public string name;         public double assetValue;         public bool previousLoanExist;         public Person(string name,             double assetValue=100000,             bool previousLoanExist = false)         {             this.name = name;             this.assetValue = assetValue;             this.previousLoanExist = previousLoanExist;         }     }

189

Chapter 9

Simplify Complex Systems Using Facades

Notice the constructor of this class. Since you have two optional parameters, you can create instances using any of the following lines: Person jack = new Person("Jack"); Person kate = new Person("Kate", 70000); Person tony = new Person("Tony", 125000, true); Now see the Asset class. This class has a method HasSufficientAssetValue to verify whether the current asset value is greater than or equal to the claim amount. Here is the Asset class:     class Asset     {         public bool HasSufficientAssetValue(Person person, double claimAmount)         {             Console.WriteLine($"Verifying whether {person.name} has sufficient asset value.");             return person.assetValue >= claimAmount ? true : false;         }     } Now see the LoanStatus class. This class has a method HasPreviousLoans to verify whether a person has an existing loan.     class LoanStatus     {         public bool HasPreviousLoans(Person person)         {             Console.WriteLine($"Verifying whether {person.name} has any previous loans.");             return person.previousLoanExist;         }     } In the method body, I have used a conditional operator. I could use an if-else chain also. You can opt for either of these to make the methods HasSufficientAssetValue and HasPreviousLoans work. 190

Chapter 9

Simplify Complex Systems Using Facades

POINT TO NOTE Also notice the simplified statement I use here: return person.previousLoanExist; instead of using the following line: return person.previousLoanExist ? true : false; I could do the same for the HasSufficientAssetValue() method inside the Asset class. I have kept both these variations so that you can be familiarized with them.

Demonstration 1 The Person, Asset, and LoanStatus classes are already shown. For simplicity, I put all three classes and the following client code into a single file. I do not repeat these classes in the following code segment.

Note  When you download the source code from the Apress website, refer to the folder “ImplementationWithoutFacade” inside “Chapter9” to see the complete program. Now, assume that a novice programmer writes the following client code. He creates a person instance, named bob, and shows whether he is eligible for a loan. It works, but is it a good solution? We’ll analyze this next. class Program     {         static void Main()         {              Console.WriteLine("***Directly interacting with the subsystems.***");            Asset asset = new Asset();            LoanStatus loanStatus = new LoanStatus();            string status = "approved"; 191

Chapter 9

Simplify Complex Systems Using Facades

           string reason = String.Empty;            bool assetValue, previousLoanExist;            // Person-1            Person bob = new Person("Bob", 5000, true);             // Starts background verification             assetValue = asset.HasSufficientAssetValue(bob, 20000);             previousLoanExist = loanStatus.HasPreviousLoans(bob);             if (!assetValue)             {                 status = "Not approved.";                 reason += "\nInsufficient balance.";             }             if (previousLoanExist)             {                 status = "Not approved.";                 reason += "\nOld loan exists.";             }             Console.WriteLine($"{bob.name}'s application status: {status}");             Console.WriteLine($"Remarks if any: {reason}");             Console.ReadKey();         }     }

Output Here is the output: ***Directly interacting with the subsystems.*** Verifying whether Bob has sufficient asset value. Verifying whether Bob has any previous loans. Bob's application status: Not approved. Remarks if any: Insufficient balance. Old loan exists. 192

Chapter 9

Simplify Complex Systems Using Facades

A  nalysis Let me ask you a few questions: •

There is only one customer at this moment. What will you do if you have two or more loan applicants? Will you repeat the background verification logic multiple times inside Main()?



Have you noticed that inside the client code you expose your background verification logic? Is this a good idea?



How would you feel if you did not need to create subsystem instances (for example, an Asset or LoanStatus instance) to know the outcome? Instead, you could assume that there is a loan approver instance that is the only point of contact that lets you know the status of an application. This would enable you to write something like the following:

Person bob = new Person("Bob", 5000,true); string approvalStatus = loanApprover.CheckLoanEligibility(bob, 20000); Console.WriteLine($"{bob.name}'s application status:{approvalStatus}"); •

In the future, if there are new criteria to get a loan, let the loan approver take responsibility for handling the situation.

B  etter Program When you consider such questions, you realize that you should search for a better solution. You can make a single point of contact (like the loan approver) to make your code cleaner and more easily maintained.

C  lass Diagram Figure 9-1 shows the class diagram for the most important parts of Demonstration 2.

193

Chapter 9

Simplify Complex Systems Using Facades

Figure 9-1.  A client directly talks to LoanApprover to know whether he can get a loan

Demonstration 2 Here is an improved version of Demonstration 1: using System; namespace UsingFacade {     class Person     {         public string name;         public double assetValue;         public bool previousLoanExist;         public Person(string name,             double assetValue=100000,             bool previousLoanExist = false) 194

Chapter 9

Simplify Complex Systems Using Facades

        {             this.name = name;             this.assetValue = assetValue;             this.previousLoanExist = previousLoanExist;         }     }     class Asset     {         public bool HasSufficientAssetValue(Person person, double claimAmount)         {             Console.WriteLine($"Verifying whether {person.name} has the sufficient asset value.");             return person.assetValue >= claimAmount ? true : false;         }     }     class LoanStatus     {         public bool HasPreviousLoans(Person person)         {             Console.WriteLine($"Verifying whether {person.name} has any previous loans.");             //return person.previousLoanExist ? true : false;             // simplified statement             return person.previousLoanExist;         }     }     class LoanApprover     {         readonly Asset asset;         readonly LoanStatus loanStatus;         public LoanApprover()         {             asset = new Asset();             loanStatus = new LoanStatus();         } 195

Chapter 9

Simplify Complex Systems Using Facades

         public string CheckLoanEligibility(Person person, double claimAmount)         {             string status = "approved";             string reason = String.Empty;             Console.WriteLine($"\nChecking the loan approval status of {person.name}.");             Console.WriteLine($"[Current asset value:{person.assetValue}," +                 $"claim amount:{claimAmount}," +                 $"existing loan?:{person.previousLoanExist}.]\n");             if (!asset.HasSufficientAssetValue(person,claimAmount))             {                 status = "Not approved.";                 reason += "\nInsufficient balance.";             }             if(loanStatus.HasPreviousLoans(person))             {                 status = "Not approved.";                 reason +="\nOld loan exists.";             }             return string.Concat(status,"\nRemarks if any:",reason);         }     }     class Program     {         static void Main()         {             Console.WriteLine("***Simplifying the usage of a complex system using a facade.***");             // Using a facade             LoanApprover loanApprover = new LoanApprover();             // Person-1             Person bob = new Person("Bob", 5000,true);             string approvalStatus = loanApprover.CheckLoanEligibility (bob, 20000); 196

Chapter 9

Simplify Complex Systems Using Facades

            Console.WriteLine($"{bob.name}'s application status: {approvalStatus}");             // Person-2             Person jack = new Person("Jack");             approvalStatus = loanApprover.CheckLoanEligibility (jack, 30000);             Console.WriteLine($"{jack.name}'s application status: {approvalStatus}");             // Person-3             Person tony = new Person("Tony", 125000,true);             approvalStatus = loanApprover.CheckLoanEligibility (tony, 50000);             Console.WriteLine($"{tony.name}'s application status: {approvalStatus}");             Console.ReadKey();         }     } }

Output Here is the output: ***Simplifying the usage of a complex system using a facade.*** Checking the loan approval status of Bob. [Current asset value:5000,claim amount:20000,existing loan?:True.] Verifying whether Bob has sufficient asset value. Verifying whether Bob has any previous loans. Bob's application status: Not approved. Remarks if any: Insufficient balance. Old loan exists.

197

Chapter 9

Simplify Complex Systems Using Facades

Checking the loan approval status of Jack. [Current asset value:100000,claim amount:30000,existing loan?:False.] Verifying whether Jack has sufficient asset value. Verifying whether Jack has any previous loans. Jack's application status: approved Remarks if any: Checking the loan approval status of Tony. [Current asset value:125000,claim amount:50000,existing loan?:True.] Verifying whether Tony has sufficient asset value. Verifying whether Tony has any previous loans. Tony's application status: Not approved. Remarks if any: Old loan exists.

Analysis Using facades, you can get the following benefit: •

You make a simplified interface for your clients.



You reduce the number of objects that a client needs to deal with.



If there are many subsystems, managing those subsystems with a facade can make communication easier.

Summary This chapter shows you how to use a facade in an application. Facades can help you develop a simplified interface for your clients dealing with many subsystems. I also discussed the difference between a facade and a template method. This chapter also reviewed the different types of facades and the pros and cons of using them in an application. It’s worth remembering the following points before you consider using a facade in an application:

198

Chapter 9

Simplify Complex Systems Using Facades



You should not assume that you can have only one facade in an application. You can use two or more of them if you find them useful.



One facade can show a different behavior than another facade. For example, you may allow or disallow direct access to the subsystems. When you force the client to create instances through a facade, you call it an opaque facade. When you also allow direct access to the subsystems, you are using a transparent facade.



If a subsystem changes, you need to incorporate the corresponding behavior in the facade layer.



Using facades, you maintain an additional layer of coding. You need to test this layer before you deliver the product to a customer. If the facade is too complex, it can produce some additional maintenance costs.

199

PART IV

The Road Ahead Part IV consists of two chapters, in which we’ll try to find out the answers to some important questions, such as the following: •

How does garbage collection (GC) work in C#?



How can we identify and tackle memory leaks in an application?



How can we decide between a static method and an instance method?



Why is understanding design patterns important?



How can we avoid anti-patterns?

In addition, we will become familiar with some terms that are commonly used in the context of software development. Each chapter contains a few Q&A sessions. A quick overview of these topics can help you to think better and program better in the future.

CHAPTER 10

Memory Management Memory management is a large concern for a developer, and it’s a very big topic to cover. This chapter aims to touch on the primary points in a simplified manner and help you understand their vitalness in programming. Simply following some design guidelines when creating an application is not enough; it is only one part of the equation. An application is truly efficient when there are no memory leaks. If a computer program runs over a long time, but fails to release memory resources that are no longer needed, you can just guess the impact of any memory leaks. Here are some common symptoms: •

A machine becomes slow over time.



A specific operation in an application takes longer to execute.



As the worst case, the application/system can crash.

A novice C# programmer often believes that the garbage collector (GC) can take care of memory management in every possible scenario. But that is not true, and unfortunately it is a common mistake. This chapter is made for this discussion and suggests you prevent memory leaks to make a better and more efficient application.

O  verview Consider a simple example. Suppose that you have an online application where users need to fill in some data and then click a Submit button. Now assume that the developers of the application mistakenly forgot to deallocate some memory that is no longer needed once the user presses the Submit button. Let us say that the application is leaking 512 bytes per click as a result of this misjudgment. You probably won’t notice any performance degradation during the initial clicks. But what happens if thousands of online users use the application simultaneously? If 100,000 users press the Submit button, we will eventually lose 48.8 MB of memory, 10 million (10,000,000) clicks leads to the loss of 4.76 GB, and so on. © Vaskaran Sarcar 2021 V. Sarcar, Simple and Efficient Programming with C#, https://doi.org/10.1007/978-1-4842-7322-7_10

203

Chapter 10

Memory Management

In short, even if a program leaks a very small amount of data for a common operation, it is quite obvious that you will see some kind of malfunction over time, such as your device crashing with a System.OutOfMemoryException, or operations in the device becoming so slow that you need to restart the application often. How fast it comes to your attention depends on the leaking rate of the application. In an unmanaged language like C++, you deallocate the memory once the intended job is completed so as to avoid memory leaks. But .NET always tries to make your programming life easier. It takes responsibility for clearing the objects that have no use after a particular point. In programming, we call them dirty objects or unreferenced objects. How does it clear the dirty objects? In .NET, the heap memory is managed. This means the common language runtime (CLR) takes care of this responsibility. In managed code, CLR’s garbage collector does this job for you, and you do not have to deallocate the managed memory. It removes the unused stuff on the heap and recollects the memory for further use. The garbage collector program runs in the background as a low-priority thread. It keeps track of the dirty objects for you. The .NET run-time can invoke this program at regular intervals to remove unreferenced or dirty objects from memory. At a given point of time, if an object has no reference, the garbage collector marks this object and reclaims the memory occupied by the object, assuming that it is no longer needed.

Note  In theory, once a local variable references an object, that object is ready for garbage collection at the earliest point at which it is no longer needed. But if you disable the optimization in debug mode, the lifetime of the object extends to the end of the block. But GC may not reclaim the memory immediately. There are various factors, such as available memory and the time since the last collection. This means an orphaned object can be released immediately, or there may be some delay.

204

Chapter 10

Memory Management

However, there is a catch. Some objects require special codes to release their resources. Here are some common examples: you open a file, perform some reading or writing, but forget to close the file. A similar kind of attention is needed when you deal with unmanaged objects, locking mechanisms, the operating system (OS) handles in your programs, and so forth. Programmers need to explicitly release those resources so as to prevent memory leaks. In general, when programmers themselves clean up (or release) the memory, you say that they dispose of the objects, but when CLR automatically releases the resources, you say that the garbage collector performs its job. The garbage collector uses the finalizers (or destructors) of the class instance to perform the final clean-up. You’ll see a discussion on them shortly.

POINTS TO REMEMBER Programmers can release resources by explicitly disposing of the objects, or CLR can automatically release resources through a garbage collection mechanism. We often refer to them as disposing and finalizing techniques, respectively.

Stack Memory vs. Heap Memory To understand the upcoming discussion, it’s important to understand the difference between stack memory and heap memory. If you know the difference, you can skip this section. Otherwise, continue reading. To execute a program, the operating system gives you a pile of memory. The program splits this into several portions for various uses. There are two major parts: one is a stack and one is a heap. The stack is used for local variables and to keep track of the current state of the program. The stack follows the Last In First Out (LIFO) mechanism. It works like a stack of frames, where one frame is placed on top of another frame. You can also think of it as a set of boxes, where one box is placed on top of another box. All local variables of a particular method can go into a single frame. At a particular moment, you can access the top frame of the stack, but you cannot access the lower frames. Once the control returns

205

Chapter 10

Memory Management

from a method, the top frame is removed from the stack and discarded. The immediate lower frame can be accessed as it becomes the top frame. This process can continue until the stack is empty. To demonstrate this, let us consider the following code: using System; class SampleStackDemo {   static void GetAnotherInt()   {     int c=3;   } static void Main()   {     int a=1;     double b=2.5;     GetAnotherInt();   } } See the following figure (Figure 10-1). I have shown you four different stages in a single snapshot. This figure shows the various stack states, which are as follows:

206



The first two lines inside the Main() method are already executed, and the third line inside the Main() method (GetAnotherInt();) starts executing. Assume that the control enters into the actual method body and passes the line int c=3; inside this method, but it does not reach the end of the method body. You can see that the stack is growing in this stage.



The next figure shows that the control comes out from GetAnotherInt(). So, c=3 is no longer on the stack.



The next figure shows that cleaning up the stack is in progress. When the control leaves Main(), both a and b variables are deleted. But following the LIFO (Last In First Out) structure, I show you the intermediate deletions one by one.

Chapter 10

Memory Management

Figure 10-1.  The different statuses of the stack memory when a program runs In short, for a stack allocation, you know that once you return from a method, the top frame is discarded and you can use the space immediately. On the other hand, the heap memory is used for object/reference types. Here, the tracking of a program state is not the concern. Instead, it focuses on storing the data. A program can easily allocate some space in the heap and start using the space to store the information.

Note Once you learn multithreaded programming, you’ll see that each thread can have its own stack, but they share the same heap space among them. For a heap, you can add or remove allocated space in any order. Here is a sample figure for your easy understanding (Figure 10-2).

207

Chapter 10

Memory Management

Figure 10-2.  A sample figure that represents a heap memory with different allocations In this case, you need to remember the allocation, and before you reuse the space, someone needs to clear the old allocation. But what happens if you forget to delete the previously allocated memory, or if you use an already created reference to point to a different object in the heap, or you make it null? These allocated memory spaces will keep increasing (which becomes garbage), and you’ll see the impact of the memory leaks. And this is the point where the garbage collector (GC) in C# helps you. Periodically, the GC checks the status and tries to help you by freeing unused spaces. Each time you create an object, the CLR allocates memory in the managed heap. It can keep allocating memory until the address space in the managed heap is available. The GC has an optimizing engine to determine when to reclaim the unused memories.

Q&A Session 10.1 I have a solution in my mind. I can allocate memory on the heap, and once my job is done, I’ll delete it immediately. This way I can prevent the garbage from growing. Is this understanding correct? Answer:

208

Chapter 10

Memory Management

Yes, the proposed solution can work, and help you prevent leaks. But this is not that easy. There are situations where the objects need to stay alive for a while. Consider an example: using an advanced printer, you simultaneously send multiple emails and faxes to different recipients. At the same time, you start printing some large documents. It is very unlikely, all the recipients receive the data at the same time, or a document with a big number of pages is printed instantly. So, an immediate deletion is not a clever solution in these scenarios. 10.2 Let us assume there is a class called Test. I understand that for the following line, Test testObj=new Test();, the space for the object will be allocated in the heap memory. But what about the reference variable? Answer: The reference variable will stay in the stack memory. Figure 10-3 depicts the scenario.

Figure 10-3.  An object reference on the stack points to the actual memory in the heap 10.3 Sometimes I wonder about these references? Are they similar to the pointers in C/C++? Answer: The concept is similar, but not the same. Before I answer your question, let me explain something further for better understanding. I already mentioned that the GC manages the heap memory for you. How does it manage this stuff? In simple terms: •

It frees up the garbage/unused spaces for you so that you can reuse the space. 209

Chapter 10



Memory Management

Secondly, it can apply the compaction technique, which means it can remove all allocated space to one side of the memory and all the free space to the other side of the memory. This results in contiguous free space that helps you to allocate a large block of memory.

The first point is important, and is the topic of this chapter. The second point is also important, because the heap may contain scattered objects (see Figure 10-2). In many situations, you may need to have a big chunk of contiguous memory, which may not be available at a particular time, though there is technically enough space in the heap. In these scenarios, the compaction helps to get enough contiguous space. These references are maintained by the garbage collector, and when this kind of shuffling is done you are not aware of it.

Note Actually, you have two different types of heap: one is a large object heap (LOH), the other one is a small object heap (SOH). The objects of sizes 85,000 bytes and above are placed in the large object heap. Usually, these are array objects. To make the discussion easy, I simply refer to the word “heap” instead of categorizing it. The SOHs are used for three different generations, which you will read about in the following section. To elaborate on these with simple figures, let us assume this is our heap (Figure 10-4). After the garbage collector’s cleanup operation, it may look like the following (white blocks are represented as free/available blocks).

Figure 10-4.  Scattered allocations in the memory before the compaction

210

Chapter 10

Memory Management

You can see that if you need to allocate five contiguous memory blocks in our heap, you cannot allocate them now, although collectively there are enough spaces. To deal with this situation, the garbage collector can apply the compaction technique, which moves all remaining objects (live objects) to one end to form one continuous block of memory. So, after compaction, it may look like Figure 10-5.

Figure 10-5.  Revised allocations in the memory after the compaction Now you can easily allocate five contiguous blocks of memory in the heap. In this way, a managed heap is different from an unmanaged heap. Here we do not need to iterate through a linked list of addresses to find spaces for new data—you can simply use the heap pointer— so, instantiating an object under .NET is faster. The online link https://docs.microsoft. com/en-us/dotnet/standard/garbage-collection/fundamentals also states the following:

Allocating memory from the managed heap is faster than unmanaged memory allocation. Because the runtime allocates memory for an object by adding a value to a pointer, it’s almost as fast as allocating memory from the stack. In addition, because new objects that are allocated consecutively are stored contiguously in the managed heap, an application can access the objects quickly. POINTS TO REMEMBER  What do I mean by an unmanaged heap? Consider a case where you personally manage the heap and are responsible for allocating and deallocating spaces. When an object is allocated in a managed heap, instead of getting the actual pointer, you get a “handle” to represent a direction to a memory address. This is helpful because the actual memory location can be changed after GC’s compaction. But for a native code (say when you use malloc() function in the C/C++ code to allocate a space), you get pointers, not handles.

211

Chapter 10

Memory Management

After the compaction, objects generally stay in the same area, so accessing them also becomes easier and faster (because page swapping is less). The compaction technique is costly, but the overall gain can be worth it. The Microsoft documentation says the following:

Memory is compacted only if a collection discovers a significant number of unreachable objects. If all the objects in the managed heap survive a collection, then there is no need for memory compaction. To improve performance, the runtime allocates memory for large objects in a separate heap. The garbage collector automatically releases the memory for large objects. However, to avoid moving large objects in memory, this memory is usually not compacted.

Note  If you are interested in further details, I encourage you to read the following .NET blog article: https://devblogs.microsoft.com/dotnet/largeobject-heap-uncovered-from-an-old-msdn-article/ Now I come to the original question. It is important how you interpret the word “pointer.” In C/C++, using a pointer, you point to an address that is nothing but a number slot in the memory. But the problem is, if you point to an invalid address, you encounter surprises! So, a pointer in an “unsafe” context is tricky. On the other hand, a reference in C# points to a valid address in the managed heap or it is null. This guarantee you get from C#. In addition, pointers are very useful because when the data moves around in the memory, you can still access those data using these references.

The Garbage Collector in Action A generational garbage collector (GC) is used to collect short-lived objects more frequently than longer-lived objects. We have three generations here: 0, 1, and 2. Shortlived objects are stored in generation 0. The longer-lived objects are pushed into the higher generations—either 1 or 2. The garbage collector works more frequently in the lower generations than in the higher ones.

212

Chapter 10

Memory Management

Once you create an object, it resides in generation 0. When generation 0 is filled up, the garbage collector is invoked. The objects that survive generation 0 garbage collection are transferred to the next higher generation—generation 1. The objects that survive garbage collection in generation 1 enter the highest generation—generation 2. The objects that survive generation 2 garbage collection stay in the same generation.When the garbage collector detects that the survival rate is too high in a generation, it increases the threshold of allocations for the generation. Finally, if it fails to allocate any further memory, you see the impact of memory leaks that you’ll learn shortly in this chapter.

Note  Sometimes you create a very large object. This kind of object goes directly to the large object heap (LOH). It is often referred to as generation 3. Generation 3 is a physical generation that's logically collected as part of generation 2. In this context, I encourage you to read the online Microsoft doc at https://docs.microsoft.com/ en-us/dotnet/standard/garbage-collection/fundamentals. I suggest you remember a 3–3 rule to remember the different phases of garbage collection and the different ways to invoke the GC.

Different Phases of Garbage Collection The following are the three different phases of garbage collection: •

Phase 1: It is the marking phase, in which the live objects are marked or identified.



Phase 2: It is the relocating phase, in which it updates the references of the objects that will be compacted in phase 3.



Phase 3: It is the compacting phase, which reclaims memory from dead (or unreferenced) objects; the compaction operation is performed on the live objects. It moves the live objects (that survived until this point) to the older end of the segment.

Different Cases of Invoking the Garbage Collector The following are three different cases of invoking the garbage collector: •

Case 1: You have low memory. 213

Chapter 10

Memory Management



Case 2: The allocated objects (in a managed heap) surpass a defined threshold limit.



Case 3: You call the System.GC() method. There are many overloaded versions of GC.Collect(). The GC is a static class and defined in the System namespace.

The following program demonstrates a simple case study for this. I have used the GetTotalMemory() method in this example. I pick the summary from Visual Studio for your immediate reference. The explanation is clear: // Summary: // Retrieves the number of bytes currently thought to be allocated. // A parameter indicates whether this method can wait for // a short interval before returning, to allow the system // to collect garbage and finalize objects. // // Parameters: //   forceFullCollection: //     true to indicate that this method can wait for garbage collection to //     occur before returning; otherwise, false. // // Returns: //     A number that is the best available // approximation of the number of bytes currently // allocated in managed memory. Similarly, you can see the descriptions of any method from Visual Studio. Here are some additional methods’ brief descriptions. I use them in the upcoming example:

214



GC.Collect(Int32) forces an immediate garbage collection from generation 0 through a specified generation. It means that when you call GC.Collect(0), the garbage collection will happen at generation 0; if you call GC.Collect(1), the garbage collection will happen both at generation 0 and generation 1, and so forth.



The CollectionCount method returns the number of times garbage collection has occurred for the specified generation of objects.

Chapter 10

Memory Management



After I invoke the GC, I invoke the WaitForPendingFinalizers() method. The method definition in Visual Studio says that this method “suspends the current thread until the thread that is processing the queue of finalizers has emptied that queue.”



C# 9.0 allows you to access whether an object is not null. So, the following block of code does not create a compile-time error: if (sample is not null){// some code}



At the time of this writing, there are five overloaded methods for Collect():

public static void Collect(); public static void Collect(int public static void Collect(int public static void Collect(int blocking); public static void Collect(int blocking, bool compacting);

generation); generation, GCCollectionMode mode); generation, GCCollectionMode mode, bool generation, GCCollectionMode mode, bool

You can see their definitions easily in Visual Studio. For your immediate reference, I present the parameters’ descriptions here: generation: It is the number of the oldest generation to be garbage collected. mode: An enumeration value that specifies whether the garbage collection is forced (System.GCCollectionMode.Default or System.GCCollectionMode.Forced) or optimized (System.GCCollectionMode.Optimized). blocking: You set it to true to perform a blocking garbage collection; false to perform a background garbage collection where possible. compacting: You set it to true to compact the small object heap; false to sweep only. The purpose of this demonstration is: •

to show you different generations of garbage collection, and



to demonstrate that an object CAN enter from one generation to the next generation if the garbage is not collected.

215

Chapter 10

Memory Management

Demonstration 1 Run the following program and go through the output and analysis: using System; namespace GCDemo {     class Sample     {         public Sample()         {             // Some code         }     }     class Program     {         public static void Main(string[] args)         {             Console.WriteLine("*** Exploring Garbage Collections.***");             try             {                 Console.WriteLine($"Maximum GC Generation is {GC. MaxGeneration}");                 Sample sample = new Sample();                 CheckObjectStatus(sample);                 for (int i = 0; i < 3; i++)                 {                     Console.WriteLine($"\n After GC.Collect({i})");                     GC.Collect(i, GCCollectionMode.Forced, false, true);                     System.Threading.Thread.Sleep(5000);                     GC.WaitForPendingFinalizers();                     ShowAllocationStatus();                     CheckObjectStatus(sample);                 }             } 216

Chapter 10

Memory Management

            catch (Exception ex)             {                 Console.WriteLine("Error:" + ex.Message);             }             Console.ReadKey();         }         private static void CheckObjectStatus(Sample sample)         {             if (sample is not null) //C# 9.0 onwards             {                 Console.WriteLine($" The {sample} object is in Generation:{GC.GetGeneration(sample)}");             }         }         private static void ShowAllocationStatus()         {             Console.WriteLine("---------");             Console.WriteLine($"Gen-0 collection count:{GC.Collection Count(0)}");             Console.WriteLine($"Gen-1 collection count:{GC.Collection Count(1)}");             Console.WriteLine($"Gen-2 collection count:{GC.Collection Count(2)}");             Console.WriteLine($"Total Memory allocation:{GC.GetTotalMemory (false)}");             Console.WriteLine("---------");         }     } }

217

Chapter 10

Memory Management

Output Here is one possible output. I have highlighted some important lines in bold. On your computer, you may see different outputs. Look over the analysis section to learn more about this difference. Maximum GC Generation is 2 The GCDemo.Sample object is in Generation: 0 After GC.Collect(0) --------Gen-0 collection count:1 Gen-1 collection count:0 Gen-2 collection count:0 Total Memory allocation:347360 --------The GCDemo.Sample object is in Generation: 1 After GC.Collect(1) --------Gen-0 collection count:2 Gen-1 collection count:1 Gen-2 collection count:0 Total Memory allocation:178984 --------The GCDemo.Sample object is in Generation: 2 After GC.Collect(2) --------Gen-0 collection count:3 Gen-1 collection count:2 Gen-2 collection count:1 Total Memory allocation:178824 --------The GCDemo.Sample object is in Generation: 2

218

Chapter 10

Memory Management

POINT TO NOTE It is possible to see the different counters if additional garbage collection happens in between these calls. In this possible output, you can see that the sample instance was not collected in any of the GC invocation calls. So, it survived and gradually moved to generation 2. The total memory allocations in this output seem to be logical because, after each GC invocation, you see that the total allocations are reducing. It may not happen in every possible output. This is because you may not allow the GC to complete its job before you show the memory status. So, to get a more consistent result, I also introduce a sleep time after I invoke the GC, and I also invoke WaitForPendingFinalizers(). This gives the GC more time to complete its job. Yes, it causes some performance penalties, but in my system, it produces a more consistent result. Based on your system configuration, you may need to vary the sleep time accordingly. Notice that I have used the following overloaded version: GC.Collect(i, GCCollectionMode.Forced, false, true). I make the third parameter false to perform a background garbage collection if possible. Note that before a garbage collection starts, all the managed threads are suspended, except the thread that invokes the GC. So, once the GC finishes its task, the other threads can start allocating spaces again. If you know the concept of multithreading, understanding the previous line is easy for you. One last point: these generations are a logical view of the GC heap. Physically, these objects reside on the managed heap, which is a chunk of memory. GC reserves this from the OS by calling VirtualAlloc. We are not going into that here, however.

Analysis This is only a sample output; it can vary in every run. If needed, you can go through the theory in the previous sections again and then try to understand how garbage collection happened. Here are some important observations: •

You can see the different generations of GC.

219

Chapter 10

Memory Management



You can see that once you called GC.Collect(2), the other generations are also called—notice that the counters are increased. Similarly, when you called GC.Collect(1), generation 1 and generation 0 both are called.



You can also see that the object that I have created was originally placed in generation 0.

D  isposing of an Object Often, a programmer needs to explicitly release some resources. Some common examples include when you work with events, locking mechanisms, file-handling operations, or unmanaged objects. There are also cases when you know that you have used a very large block of memory that is not necessary after a certain point of execution. These are some examples where you want to release the memory or resources to improve the performance of your system.

Note An unmanaged object is not controlled by .NET. A common example is when you wrap OS resources such as database connections or network connections. In .NET, you have an IDisposable interface with a Dispose() method. When a programmer wants to release resources, he can override this Dispose() method. This is a recommended practice, because you are very much aware of when you are about to release the memory. Figure 10-6 shows a snapshot from Visual Studio that says you can release unmanaged resources using this method.

220

Chapter 10

Memory Management

Figure 10-6.  IDisposable interface in .NET

Finalize vs. Dispose Each class can have only one finalizer (often called a destructor), which cannot be overloaded or inherited. It does not have a modifier, and it does not take any parameter. You cannot call a finalizer directly. It is invoked automatically. Here is a sample example that shows a finalizer or a destructor: class Sample {    ~Sample() // finalizer    {      // Cleanup statements...    } } If you compile this code segment and then open the IL code, you notice something like the following: .method family hidebysig virtual instance void         Finalize() cil managed {   .override [mscorlib]System.Object::Finalize   // Code size       13 (0xd) 221

Chapter 10

Memory Management

  .maxstack  1   IL_0000:  nop   .try   {     IL_0001:  nop     IL_0002:  leave.s    IL_000c   }  // end .try   finally   {     IL_0004:  ldarg.0     IL_0005:  call       instance void [mscorlib]System.Object::Finalize()     IL_000a:  nop     IL_000b:  endfinally   }  // end handler   IL_000c:  ret } // end of method Sample::Finalize

Note You can use an IL disassembler to see the IL code. I often use ildasm.exe, which is automatically available in Visual Studio. To use this tool, you can follow these steps: Open Developers Command Prompt for Visual Studio ➤ Type ildasm (You will see a new window pop up) ➤ Drag a .dll or an .exe to this window ➤ Now expand/click the code elements. You can learn more about this tool from this online link: https://docs.microsoft.com/en-­us/dotnet/framework/ tools/ildasm-­exe-­il-­disassembler It is because a finalizer call implicitly translated to: protected override void Finalize() {     try     {         // Cleanup statements...     }

222

Chapter 10

Memory Management

    finally     {         base.Finalize();     } } This method is called recursively for all instances in an inheritance chain, and the direction of the call is from the most special to the least special.

Note  Microsoft recommends to not use empty finalizers, because an entry is made in the Finalize queue for each finalizer. When a finalizer is called, the GC starts processing this queue. So, if the finalizer is empty, you introduce an unnecessary performance penalty. Let us look at a program where you see the presence of both a finalizer and a Dispose() method. Before you run this program, let me tell you a few things: •

The static class GC is defined in the System namespace.



This class has a method, called SuppressFinalize(). If you pass the current object in the GC.SuppressFinalize() method, the finalize method of the current object is not invoked.



I want to show you a destructor invocation in .NET 5 or .NET 6. In .NET Framework, it is very easy. Once you exit the program, it is called automatically. But a different logic is implemented in the .NET core platform (or .NET 5 or .NET 6). This is why I introduce another class, called A, and initialize a Sample object inside the constructor. I also do not use any Sample reference inside Main() before I invoke the GC. This helps the GC to analyze whether the Sample object is no longer needed and then collect the garbage. A similar logic can be implemented to mimic this behavior in .NET 5/.NET 6/.NET core platform.

223

Chapter 10

Memory Management

POINT TO REMEMBER Ideally, unless it is very much required, you do not want to write code in the finalizer. Instead, you may prefer to use the Dispose() method to release the unmanaged resources and avoid memory leaks.

Demonstration 2 Run the following program now and follow the output. Then go through the analysis. You need to understand an important design change in the .NET platform. using System; namespace DisposeExample {     class Sample : IDisposable     {         public void SomeMethod()         {             Console.WriteLine("Sample's SomeMethod is invoked.");         }         public void Dispose()         {             // GC.SuppressFinalize(this);             Console.WriteLine("Sample's Dispose() is called");             // Release unmanaged resource(s) if any         }         ~Sample()         {             Console.WriteLine("Sample's Destructor is called.");         }     }     class A : IDisposable     {         public A() 224

Chapter 10

Memory Management

        {             Console.WriteLine("Inside A's constructor.");             // C#8 onwards it works.             // using Sample sample = new Sample();             // sample.SomeMethod();             using (Sample sample = new Sample())             {                 sample.SomeMethod();             }         }         public void Dispose()         {             // GC.SuppressFinalize(this);             Console.WriteLine("A's Dispose() is called.");             // Release any other resource(s)         }         ~A()         {             Console.WriteLine("A's Destructor is Called.");         }     }     class Program     {         static void Main(string[] args)         {             Console.WriteLine("*** Exploring the Dispose() method.***");             A obA = new A();             obA = null;             Console.WriteLine("GC is about to start.");             GC.Collect();             GC.WaitForPendingFinalizers();             Console.WriteLine("GC is completed.");             Console.ReadKey();         }     } } 225

Chapter 10

Memory Management

Output Here is the output when I use .NET5, .NET 6, or .NET Core 3.1: *** Exploring the Dispose() method.*** Inside A's constructor. Sample's SomeMethod is invoked. Sample's Dispose() is called. GC is about to start. Sample's Destructor is called. GC is completed.

Analysis From this output, you can notice the following points: •

The Sample class object’s Dispose() and finalizer methods are both called.



The statement GC.SuppressFinalize(this); is commented in the dispose() method of the Sample class. This is why the destructor of the Sample instance was called too. If you enable/uncomment this statement, the finalizer of the Sample instance will not be called.



The A object’s finalizer method is not called yet.

When I execute the same program in the .NET Framework 4.7.2, I can see an additional line toward the end that says that A class object’s destructor is also called in this case. Here is the output: *** Exploring the Dispose() method.*** Inside A's constructor. Sample's SomeMethod is invoked. Sample's Dispose() is called. GC is about to start. Sample's Destructor is called. A's Destructor is Called. GC is completed.

226

Chapter 10

Memory Management

Note  I raised a ticket with Microsoft regarding the difference in output in NET Framework and .NET Core. If you are interested to know about this discussion, you can refer to the link: https://github.com/dotnet/docs/issues/24440 Microsoft believes that it is an expected behavior in.NET Core/.NET 5/. NET 6 applications. There are different opinions too. In this context, I refer to the Microsoft documentation (https://docs.microsoft. com/en-us/dotnet/csharp/programming-guide/classes-and-structs/destructors) that says: The programmer has no control over when the finalizer is called; the garbage collector decides when to call it. The garbage collector checks for objects that are no longer being used by the application. If it considers an object eligible for finalization, it calls the finalizer (if any) and reclaims the memory used to store the object. In .NET Framework applications (but not in .NET Core applications), finalizers are also called when the program exits. The explanation I got for this is: finalizers could produce a deadlock that would prevent a program from exiting. Therefore, the code to run finalizers on exit was relaxed further. You can refer to our discussion in the previous link. This online link https://github.com/dotnet/docs/issues/17463 describes the issue in depth. Before you move onto the topic of memory leaks in detail, let us review in the following Q&A session.

Q  &A Session 10.4 How can we call destructors? Answer: You cannot call the destructor. The garbage collector takes care of that responsibility. 10.5 What is a managed heap? And how can you free up a resource? Answer: When CLR initializes the garbage collector, it allocates a segment of memory to store and manage the objects. This memory is known as the managed heap. In general, Finalize() (or the destructor of the object) is invoked to clean up the memory. Hence, you can provide the destructor to free up unreferenced resources held by our objects. In that case, you need to override the Object class’s Finalize() method. 227

Chapter 10

Memory Management

Normally, a programmer tries to use the Dispose() method to release unmanaged resources. To optimize the performance, he may suppress the finalizer call on an object if he wants. In this context, you may see a dispose pattern something like the following:   class Sample : IDisposable     {         protected virtual void Dispose(bool disposing)         {             if( disposing)         {             // Some code to release managed resources.         }         public void Dispose()         {             Dispose( true);             GC.SuppressFinalize(this);         }         ~Sample().         {             Dispose(false);         }         // Some code         }     }

Note Notice that the disposing parameter is false when called from a finalizer. But it is true when you call it from the IDisposable.Dispose method. In other words, it is true when it is deterministically called and false when it is non-deterministically called. This follows Microsoft’s programming guidelines. 10.6 When does the garbage collector call the Finalize() method? Answer: We never know. It may call it instantly when an object is found with no references, or later when the CLR needs to reclaim some memory. But you can force the garbage collector to run at a given point by calling GC.Collect(), which has many overloaded versions. You have seen two different usages already, when I used GC.Collect(Int32) and GC.Collect() in previous demonstrations. 228

Chapter 10

Memory Management

10.7 Finalizers are called automatically when the program ends in the .NET Framework. But this is not the case in .NET Core or .NET 5 or NET 6. What is the reason behind this? Answer: In response to my ticket at https://github.com/dotnet/docs/issues/24440, the answer is summarized as: finalizers could produce a deadlock that can prevent a program from exiting. Therefore, the code to run finalizers on exit was relaxed further. Microsoft believes that it is expected behavior in .NET Core, .NET 5, and .NET 6 applications. 10.8 When should we invoke the GC.Collect()? Answer: I already mentioned that invoking GC is generally a costly operation. But in some special scenarios, you may feel certain that if you invoke GC, you’ll gain some significant benefits. Such an example may arise after you dereference a large number of objects in the code. Another common example is when you try to find memory leaks through some common operations, such as executing a test repeatedly to find leaks in the system. After each of these operations, you may try to gather different counters to analyze memory growth and to get the correct counters. I’ll discuss memory leak analysis shortly.

POINTS TO REMEMBER When we see the use of the IDisposable interface, we assume that the programmer will call the Dispose() method correctly. Some experts suggest you have a destructor also as a precautionary measure. It can help when a call to Dispose() is missed. Remember Microsoft’s philosophy (see https://docs.microsoft.com/en-us/dotnet/standard/ garbage-collection/implementing-dispose): To help ensure that resources are always cleaned up appropriately, the Dispose method should be idempotent, such that it is callable multiple times without throwing an exception. Furthermore, subsequent invocations of Dispose should do nothing.

229

Chapter 10

Memory Management

10.9 Why are you using “using” statements in the previous demonstration (Demonstration 2)? Answer: C# provides special support in this context. You can use the using statement to reduce your code size and make it more readable. It is a syntactic shortcut for the try/ finally block. To verify this, you can see the IL code for the A’s constructor that I used in Demonstration 2. I present this here, and make some bold for your reference: .method public hidebysig specialname rtspecialname         instance void  .ctor() cil managed {   // Code size       48 (0x30)   .maxstack  1   .locals init (class DisposeExample.Sample V_0)   IL_0000:  ldarg.0   IL_0001:  call       instance void [System.Runtime]System.Object::.ctor()   IL_0006:  nop   IL_0007:  nop   IL_0008:  ldstr      "Inside A's constructor."   IL_000d:  call       void [System.Console]System.Console::WriteLine (string)   IL_0012:  nop   IL_0013:  newobj     instance void DisposeExample.Sample::.ctor()   IL_0018:  stloc.0   .try   {     IL_0019:  nop     IL_001a:  ldloc.0     IL_001b:  callvirt   instance void DisposeExample.Sample::SomeMethod()     IL_0020:  nop     IL_0021:  nop     IL_0022:  leave.s    IL_002f   }  // end .try   finally   {     IL_0024:  ldloc.0 230

Chapter 10

Memory Management

    IL_0025:  brfalse.s  IL_002e     IL_0027:  ldloc.0     IL_0028:  callvirt   instance void [System.Runtime]System.IDisposable:: Dispose()     IL_002d:  nop     IL_002e:  endfinally   }  // end handler   IL_002f:  ret } // end of method A::.ctor 10.9 Can I directly allocate spaces in generation 1 or generation 2? Answer: No. A user code can allocate spaces in generation 0 or LOH only. It is the GC’s responsibility to promote an object from generation 0 to generation 1 (or generation 2).

Memory Leak Analysis How can you detect leaks? There are many tools for this purpose. For example, windbg. exe is a common tool to find memory leaks in a large application. Apart from this, you can use other graphical tools, like Microsoft’s CLR Profiler, SciTech’s Memory Profiler, Red Gate’s ANTS Memory Profiler, and so forth to find the leaks in your system. Many organizations have a company-specific memory leak tool to detect and analyze leaks. In my previous organization, our experts developed such a tool. It is a wonderful tool. I was fortunate because I could use it and learned many interesting things about memory leaks. In the latest editions of Visual Studio, there is a diagnostic tool to detect and analyze memory leaks. It is very user-friendly, and you can take various memory snapshots. Markers in the tool indicate garbage collector activities. This tool is very useful and effective: you can analyze the data in real-time while the debugging session is active. The spikes in the graph draw your attention immediately. The following program shows you a sample demonstration.

231

Chapter 10

Memory Management

Note  I assume that you know how to use events in an application. A detailed discussion of events and delegates is beyond the scope of this book. I have discussed delegates, events, and other topics in detail in my other books— Interactive C# and Getting Started with Advanced C#— which are published by Apress. The first book shows how to use the diagnostic tools as well as Microsoft’s CLR Profiler to analyze memory leaks. The second one discusses delegates and events in depth. So, if you are interested, you can have a look at these books. Here, I have added some supportive comments to help you understand the code a little better. I acknowledge that it is easy to find the problem in this program. But my core intention is to show you how to analyze leaks using the diagnostic tool. Before you run this application, ensure that you enable the option to launch the Diagnostic Tools, as shown in Figure 10-7. In Visual Studio IDE, you can see this option in Tools ➤ Option ➤ Debugging ➤ General.

Figure 10-7.  Enabling “Diagnostic Tools while debugging” option in Visual Studio

232

Chapter 10

Memory Management

Demonstration 3 Here is the complete demonstration. Inside Main(), you see two methods: one to register the event and one to unregister the event. You can see that, by mistake, I have registered too many events, and I have unregistered only one of them. These remaining unregistered events are causing the leaks in this application. using System; namespace MemoryLeakDemo1 {     delegate void IdChangedHandler(object sender, IdChangedEventArgs eventArgs);     class IdChangedEventArgs : EventArgs     {         public int IdNumber { get; set; }     }     class Sender     {         public event IdChangedHandler IdChanged;         private int Id;         public int ID         {             get             {                 return Id;             }             set             {                 Id = value;                 // Raise the event                 OnMyIntChanged(Id);             }         }

233

Chapter 10

Memory Management

        protected void OnMyIntChanged(int id)         {             if (IdChanged != null)             {                 // As suggested by compiler:                 // It is the simplified form of the following lines:                 // IdChangedEventArgs idChangedEventArgs = new                 // IdChangedEventArgs();                 // idChangedEventArgs.IdNumber = id;                IdChangedEventArgs idChangedEventArgs =                 new IdChangedEventArgs                 {                     IdNumber = id                 };                 IdChanged(this, idChangedEventArgs);             }         }     }     class Receiver     {         public void GetNotification(object sender, IdChangedEventArgs e)         {             Console.WriteLine($"Sender changed the id to:{e.IdNumber}");         }     }     class Program     {         static void Main()         {             Console.WriteLine("***Creating custom events and analyzing memory leaks.***");             Sender sender = new Sender();             Receiver receiver = new Receiver();             RegisterNotifications(sender, receiver); 234

Chapter 10

Memory Management

            UnRegisterNotification(sender, receiver);             Console.ReadKey();         }          private static void RegisterNotifications(Sender sender, Receiver receiver)         {             for (int count = 0; count < 10000; count++)             {                 // Registering too many events.                 sender.IdChanged += receiver.GetNotification;                 sender.ID = count;             }         }          private static void UnRegisterNotification(Sender sender, Receiver receiver)         {             // Unregistering only one event.             sender.IdChanged -= receiver.GetNotification;         }     } } I run this program and take different snapshots. Here, I present you a screenshot of the Diagnostic Tools window (Figure 10-8); it includes five different snapshots to analyze memory usage at a given point in time. It is a big snapshot, so move to the next page.

235

Chapter 10

Memory Management

Snapshots from Diagnostic Tools

Figure 10-8.  Different snapshots are taken using the diagnostic tools in Visual Studio

236

Chapter 10

Memory Management

Let’s analyze the difference (Objects (Diff )). For example, the fourth row shows the objects count increased by 171 compared to the previous snapshot. If you hover your mouse on this, it will tell that you can open the heap diff view for the selected snapshot sorted by object count. Let’s click on this link. I can see what’s shown in Figure 10-9.

Figure 10-9.  The object count difference in a particular snapshot We can see how the heap size is growing over time. Notice that, by mistake, I am registering an event repeatedly inside the for loop in this code: sender.IdChanged += receiver.GetNotification; Similarly, I could show you the leak using Microsoft’s CLR Profiler. But showing the usage of different tools is not the aim of the chapter. Instead, you can prevent memory leaks using any tool you like. Since the diagnostic tool is already available in the latest editions of Visual Studio, I did not want to miss the opportunity to show you its usage.

237

Chapter 10

Memory Management

Catching a memory leak requires expertise because it is not very easy. In the previous demonstration, our program has a few methods, which is why it is easy to catch the leak. But think about some typical scenarios: •

You use third-party code and the leak is there. But you cannot immediately find it, because you do not have access to the code.



The leak is revealed when some specific code path is followed. If the test team misses the path, it is hard to find the leak.



A dedicated memory leak suite maintenance may need a separate test team. Also, you cannot include all the regression tests in the memory leak suite. Running a test multiple times and gathering those counters are both time-consuming and resource-consuming activities. So, it is recommended that you shuffle the test cases often and run your memory leak test suite.



When a new bug fix takes place, a test team verifies the fix using test cases. Now you need to ask them whether those tests are already included in the memory leak test suite. If not, you need to include them. But if multiple fixes come in on one day (say 10 or more), it may not be possible to immediately include all the tests in your memory leak suite, for various reasons (for example, you may have resource constraints). Also, since you get to see the result much later and in between new fixes’ being entered into the main codebase, it is very hard to catch an earlier leak.

Summary Memory management is an important topic. This chapter gives you a quick overview, but still, it is a big chapter! A discussion of the importance of memory leaks was followed by a look at how memory is managed in C#. I started the discussion with the two different types of memory in C#: stack memory and heap memory. Then I discussed the garbage collector (GC) in C#. You saw different phases of garbage collection and learned different cases in which a GC can start its operation.

238

Chapter 10

Memory Management

Later you learned about disposing of an object programmatically. You saw a discussion on the dispose method vs. the finalize method. And in this case, you saw how .NET Framework shows a different behavior that .NET Core, .NET 5, or .NET 6. I raised a ticket to discuss this difference with experts in Microsoft, and you can see the discussion in https://github.com/dotnet/docs/issues/24440. In the final part, I showed you the usage of the diagnostic tools in Visual Studio and analyzed a memory leak using events in C#. In short, this chapter answered the following questions: •

How is heap memory different than stack memory?



What is garbage collection (GC)? How does it work in C#?



What are the different GC generations?



What are the different ways to invoke the garbage collector?



How can we force GC to invoke?



How is disposing different from finalizing in C#?



What is a memory leak?



What are the probable causes of memory leaks?



How can we use the Dispose() method effectively to collect memory?



How do we use memory leak analysis with Visual Studio’s diagnostic tools?

239

CHAPTER 11

Leftover Discussions This is the last chapter of the book. Here you will see some interesting discussions and learn some common terminology. Sometimes it’s ok to bend well-accepted rules for your application if it performs well, but when you keep coding and developing applications, you’ll find that the experts’ suggestions have great value. If you follow their advice, you’ll understand that a simple choice can have a big impact in the long run. This chapter discusses some of these topics in brief.

Static Method or Instance Method? A static method is easy to use. A novice programmer may think that it does not matter much whether he uses a static method or an instance method in his program. He knows that he can call a method without instantiating an object. He loves this. He is further impressed when he sees some static utility methods that are extremely helpful. But an experienced programmer often finds it difficult to understand whether he should use a static method or not. In each possible design, he may ask: which is better? The short answer is that there is no universal rule. I believe that it purely depends on the application you use. Let us verify the facts.

R  ecap Can you remember the simple factory (Demonstration 1) in Chapter 6? You saw the code earlier. For your immediate reference, I provide a screenshot from Visual Studio. Notice the arrow tip in Figure 11-1.

© Vaskaran Sarcar 2021 V. Sarcar, Simple and Efficient Programming with C#, https://doi.org/10.1007/978-1-4842-7322-7_11

241

Chapter 11

Leftover Discussions

Figure 11-1.  The AnimalFactory class from demonstration 1 in Chapter 6 If you investigate this, you will see the following message:

This is a partial snapshot; I am expanding the full message for you: CA1822 Member 'CreateAnimal' does not access instance data and can be marked as static. It also keeps saying: Active Members that do not access instance data or call instance methods can be marked as static. After you mark the methods as static, the compiler will emit non-virtual call sites to these members. This can give you a measurable performance gain for performance-sensitive code.

242

Chapter 11

Leftover Discussions

Note I can see this suggestion when I set the target framework to .NET 5 or .NET 6. But it is not visible when I use the target framework .NET 3.1. I did not take this suggestion in Chapter 6. The reason is obvious: •

You cannot mark a static method with virtual or abstract keywords.



As a result, you cannot override a static method. So, you cannot use the override keyword either.



When you cannot redefine a method using the override keyword, you do not get a polymorphic behavior.

In Chapter 6, I enhanced the initial implementation and deferred some responsibilities to subclasses because I wanted to implement the polymorphic behavior. When you think that you may need to do the same for your application, it’s better to make the method non-static. Let me summarize the key points for you: •

If you use a method that can get all information from its parameters and does not operate on any instance of a class, you can make the method static. For example, look into the following MyUtility class with static method ShowGreaterNumber(): class MyUtility     {         public static double ShowGreaterNumber(             double firstNumber, double secondNumber)         {              return firstNumber >= secondNumber ? firstNumber : secondNumber;         }     } It makes sense to me to use it as MyUtility. ShowGreaterNumber(24.7, 75.2) so as to print the greater number between 24.7 and 75.2. It is unnecessary to make an instance of MyUtility before you print the maximum between 243

Chapter 11

Leftover Discussions

24.7 and 75.2. You can refer to the in-built Math class to get an idea of good static methods. For example, using Math.Max(2,3) you can get 3, or using Math.Abs(-2.52) you can get 2.52. •

If you do not want to see a polymorphic behavior or you only care about the performance of your application, you can consider making your method static.



Sometimes you see a partial code and then feel that a static method makes more sense to you. But you have seen that it is quite possible that you will need to enhance your program in the future, and you will need polymorphism to make your code flexible. Whenever you are in doubt, choose a non-static method over its counterpart (i.e., a static method).

Learn Design Patterns Let’s have a tour with a time machine. Using this machine let me take you to the early days of software development to understand a common problem in those days: There is no standard to instruct the developers on how to design an application. We are unique creatures. So, each corporate team follows its own style of coding. A new member joins such a team. Understanding the current architecture is a gigantic task for this member. So, he is continually seeking help from the senior members of the team and requesting them to explain the existing architecture. He keeps on asking them: why do you follow this particular design in this code segment? The experienced developer answers his question. He also explains why the common alternatives were not considered in a previous team meeting. He also suggests the new member reuse the existing construct to reduce future development efforts. You ask: what is the problem with this? Actually, there is no problem, this is standard practice, even in today’s world. But think from a different perspective: let’s say the experienced guy tells the new member: we follow the facade pattern for this code segment, or we follow the singleton pattern in that code segment. If the new joiner already knows about these patterns of coding, how easy will his learning be? Not only this- since he knows these styles of coding, following a known pattern, it’s easy for him to contribute to the team much quickly. I hope you get some idea about the importance of knowing some standard patterns! 244

Chapter 11

Leftover Discussions

The software design patterns address this kind of issue and provide a common platform for all developers. You can think of them as the recorded experience of experts in the field. These patterns were originally intended to be applied in object-oriented designs with the intention of reuse.

Brief History of Design Patterns The original idea of design patterns came from building architect Christopher Alexander, a professor of Berkeley. He faced many problems that were similar in nature. So, he tackled them with similar kinds of solutions.

“Each pattern describes a problem, which occurs over and over again in our environment, and then describes the core of the solution to that problem, in such a way that you can use this solution a million times over, without ever doing it the same way twice.” —Christopher Alexander His original ideas were for buildings being constructed within a well-planned town. Later these concepts entered into the software engineering community. This community started believing that though these patterns were described for buildings and towns, the same concepts could be applied in object-oriented design. So, they substituted the original concepts of walls and doors with objects and interfaces. The idea was the same: you can apply a known solution to a common problem. The concepts started gaining popularity through leading-edge software developers like Ward Cunningham and Kent Beck. In 1994, the idea of design patterns entered into the mainstream of object-oriented software development through an industry conference called Pattern Languages of Program Design (PLoP) on design patterns. It was hosted by the Hillside Group, and Jim Coplien’s paper “A Development Process Generative Pattern Language” is a famous one in this context. In 1994, Erich Gamma, Richard Helm, Ralph Johnson, and John Vlissides published the book Design Patterns: Elements of Reusable Object-Oriented Software (Addison-­ Wesley, 1994). With the launch of this classic textbook, the idea of design patterns became extremely popular. In this book, they introduced 23 design patterns for software development. These authors became known as the Gang of Four. We often refer to them as the GoF. The patterns that they talked about were developed by common experiences of software developers. 245

Chapter 11

Leftover Discussions

It is important to note that the GoF discussed the design patterns in the context of C++. But C# 1.0 was released in 2002, and then it went through various changes. It grew rapidly and secured its rank among the world’s top programming languages within a short period, and in today’s market it is always in high demand. At the time of this writing, C# 9.0 is available with Visual Studio 2019. Since the concepts of design patterns are universal, they are always valuable. So, exercising the fundamental design patterns always makes you a better programmer and helps you to “upgrade” yourself. Here are some important points to remember: •

A design pattern describes a general reusable solution to software design problems. The basic idea is that while developing software, you can solve similar kinds of problems with similar kinds of solutions. The proposed solutions were tested over a long period.



Patterns are actually templates. They suggest to you how to solve a problem. A good understanding of patterns can help you to implement the best possible design much faster.



From the OOP perspective, these patterns are descriptions of how to create objects and classes and customize them to solve a general design problem in a particular context.

Each of the 23 GoF design patterns focuses on a particular object-oriented design. Each of them can describe the consequences and tradeoffs of use. The GoF categorized these 23 patterns based on their purposes, as shown next. A. Creational Patterns These patterns abstract the instantiation process. You make the systems independent from how the objects are composed, created, and represented. Here you ask: “Where should I place the ‘new’ keyword in my application?” This decision can help you determine the degree of coupling of your classes. The following five patterns belong to this category:

246



Singleton Pattern



Prototype Pattern



Factory Method Pattern



Builder Pattern



Abstract Factory Pattern

Chapter 11

Leftover Discussions

B. Structural Patterns Using these patterns, you combine classes and objects to form a relatively large structure. Normally you use inheritance or composition to group different interfaces or implementations. In Chapter 7, you saw that preferring object composition over inheritance (and vice versa) can affect the flexibility of your software. The following seven patterns fall into this category: •

Proxy Pattern



Flyweight Pattern



Composite Pattern



Bridge Pattern



Facade Pattern



Decorator Pattern



Adapter Pattern

C. Behavioral Patterns These patterns focus on algorithms and the assignment of responsibilities among objects. Here you concentrate on the objects’ communication and their interconnection. The following 11 patterns fall into this category: •

Observer Pattern



Strategy Pattern



Template Method Pattern



Command Pattern



Iterator Pattern



Memento Pattern



State Pattern



Mediator Pattern



Chain of Responsibility Pattern



Visitor Pattern



Interpreter Pattern 247

Chapter 11

Leftover Discussions

The GoF made another classification based on scope, namely whether the pattern primarily focuses on the classes or its objects. Class patterns deal with classes and subclasses. They use inheritance mechanisms, so these are static at compile time. Object patterns deal with objects that can change at run-time. So, object patterns are dynamic. For a quick reference, you can refer to the following table that was introduced by GoF:

Here Is the Good News! You have already implemented some patterns! Not only that, you actually learned at least one pattern from each category. Part III of this book helps you understand them:

248



In Chapter 6, you learned about the factory method pattern. In fact, you also learned the simple factory pattern, which is the foundation of this pattern.



In Chapter 7, you learned the decorator pattern.



In Chapter 8, you learned about the template method pattern.

Chapter 11

Leftover Discussions



In Chapter 9, you learned about the facade pattern.



Wait! There is one more for you: Chapter 2 in Part I creates a foundation for the strategy pattern.

These patterns are very common in C# applications. Congratulations! You are on the right path.

Q  &A Session 11.1 What are the differences between class patterns and object patterns? Answer: In general, class patterns focus on static relationships, while object patterns focus on dynamic relationships. As the name suggests, class patterns focus on classes and their subclasses, and object patterns focus on the objects’ relationships. The following table shows the summarized content discussed in GoF’s famous book: Class Patterns

Object Patterns

Creational

Can defer object creation to its subclasses

Can defer object creation to another object

Structural

Focuses on the composition of classes (primarily uses the concept of inheritance)

Focuses on the different ways of the composition of objects

Behavioral

Describes the algorithms and execution flows. They also use an inheritance mechanism.

Describes how different objects can work together and complete a task.

11.2 Can I combine two or more patterns in an application? Answer: Yes, in real-world scenarios, this type of activity is common.

249

Chapter 11

Leftover Discussions

11.3 Do these patterns depend on a particular programming language? Answer: Programming languages can play an important role. But the basic ideas are the same; patterns are just like templates and give you some idea in advance of how to solve a problem. Instead of just any object-oriented programming language, suppose you have chosen some other language like C. In that case, you may need to implement the core object-oriented principles such as inheritance, polymorphism, encapsulation, abstraction, and so on. So, the choice of a particular language is important because it may have some specialized features that can make your life easier. 11.4 Should I consider common data structures like arrays and linked lists as different design patterns? Answer: The GoF clearly excludes those, saying that “they are not complex, domain-specific designs for an entire application or subsystem.” They can be encoded in classes and reused as-is. So, they are not your concern in this context. 11.5 If no particular pattern is 100% suitable for my problem, how should I proceed? Answer: The infinite number of problems cannot be solved with a finite number of patterns, for sure. But if you know these common patterns and their tradeoffs, you can pick a close match. Lastly, no one prevents you from using your pattern for your own problem. But you have to tackle the risk and need to think about your return on investment. Remember that the world is always changing, and new patterns keep evolving. To understand the necessity of a new pattern, you may also need to understand why an old/ existing pattern is not enough to fulfill the requirement. These patterns attempt to make a solid foundation for you. These concepts can help you move smoothly in your professional life.

Avoid Anti-patterns Design patterns can help you make better applications. But often people misuse them and cause anti-patterns. An attractive solution often can cause a serious problem. Here is a common example: a developer implements a quick fix without analyzing the potential pitfalls to meet a delivery schedule. Now think about what will happen to the company’s reputation if a customer finds a big bug due to that quick fix. 250

Chapter 11

Leftover Discussions

Anti-patterns alert you in similar situations and help you to take precautionary measures. They remind you of the proverb: Prevention is better than the cure.

POINTS TO REMEMBER Anti-patterns not only warn about common mistakes but also suggest better solutions. Some of these solutions may not be attractive at the beginning, but in the long run, they save your time, effort, and reputation.

Brief History of Anti-patterns Undoubtedly, design patterns have helped (and are still helping) millions of programmers. Gradually, however, people started noticing negative impacts from the overuse of these patterns. For example: Many developers wanted to show their expertise without truly evaluating the consequences of using these patterns in their specific domains. As an obvious side-effect, patterns were implanted in the wrong context, low-­ quality software was produced, and ultimately there were big penalties for them or their organizations. The software industry needed to focus on the negative consequences of similar kinds of mistakes, and eventually the idea of anti-patterns evolved. Many experts started contributing to this field, but the first well-formed model came through Michael Akroyd’s presentation entitled, “AntiPatterns: Vaccinations against Object Misuse.” It was the antithesis of the GoF’s design patterns. Wikipedia says the term was coined in 1995 by computer programmer Andrew Koenig. The term “anti-pattern” became popular with the authors William J. Brown,Raphael C. Malveau,Hays W. McCormick III,Thomas J. Mowbray with their famous book Anti Patterns: Refactoring Software, Architectures, and Projects in Crisis (Robert Ipsen/ Wiley,1998). Later, Scott Thomas joined their group. They say the following:

“Because AntiPatterns have had so many contributors, it would be unfair to assign the original idea for AntiPatterns to a single source. Rather, AntiPatterns are a natural step in complementing the work of the design pattern movement and extending the design pattern model.”

251

Chapter 11

Leftover Discussions

Examples of Anti-patterns These are some examples of anti-patterns and the concepts/mindsets behind them: •

Over Use of Patterns: Developers may try to use patterns at any cost, regardless of whether it is appropriate or not.



God Class: A big object that controls almost everything with many unrelated methods.



Not Invented Here: I am a big company and I want to build everything from scratch. Though there is a library that is developed by a small company, I’ll not use that. I will make everything on my own, and once it is developed, I’ll use my brand value to announce:

“Hey Guys, we are here to provide you the ultimate library to fulfill your every need.”

252



Zero Means Null: A programmer may use some special numbers, such as -1 or 999 (or anything similar), to represent an inappropriate integer value. A similar example can be seen when he treats something like “09/09/9999” as a null date in an application. In these cases, if the user needs to have these values, he will not get those.



Golden Hammer: Mr. X believes that the technology T is always best. So, if he needs to develop a new system (that demands new learning), he will still prefer T even if it is inappropriate. He thinks, “I am old enough and quite busy. I do not need to learn any more technology if I can somehow manage it with T.”



Shoot the Messenger: You believe that the tester “John” always finds hard defects for you because he does not like you. You say that you’re already under pressure and the program deadline is approaching. So, you do not want him to involve himself in this crucial stage so as to avoid more defects.



Swiss Army Knife: A company target for a product that can serve every need of a customer. Or, imagine that a company tries to make a drug that can cure all illnesses. Or, someone wants to design software that can serve a wide range of customers with varying needs. It does not matter for him how complex the interface is.

Chapter 11

Leftover Discussions



Copy and Paste Programming: I need to solve a problem, but I already have a piece of code to deal with a similar situation. So, I can take a copy of the old code that is currently working, and then I can start modifying it if required. But when you start from existing copy, you essentially inherit all the potential bugs associated with it. Also, if the original code needs to be modified in the future, you need to implement the modification in multiple places. This approach violates the Don’t Repeat Yourself (DRY) principle.



Architects Don’t Code: I am an architect. My time is valuable. I’ll only show paths or give a great lecture on coding. There are enough implementers who should implement my idea. Architects Play Golf is also a sister of this antipattern.



Disguised Links and Ads: This comes from a mindset that fools the users and earns revenue when they click on a link or an advertisement. Often the customer does not get what he/she really wants. It is often referred to as the Dark Patterns.



Management by Numbers: Someone believes that a higher number of commits, higher number of lines of code, or higher number of defect fixing, etc. are signs of a great developer.

“Measuring programming progress by lines of code is like measuring aircraft building progress by weight.” —Bill Gates

Types of Anti-patterns Anti-patterns can belong to different categories. Even a typical anti-pattern can belong to more than one category. Here are some common classifications: •

Architectural Anti-patterns: Swiss Army Knife is an example in this category.



Development Anti-patterns: God Class and Overuse of Patterns are examples in this category.

253

Chapter 11

Leftover Discussions



Management Anti-patterns: Shoot the Messenger can fall into this category.



Organizational Anti-patterns: Architects Don’t Code and Architects Play Golf can belong in this category.



User Interface Anti-patterns: Example includes Disguised Links and Advertisements.

POINTS TO NOTE • You can learn about various anti-patterns from different websites/sources. For example, the following Wikipedia link talks about various anti-patterns: https://en.wikipedia.org/wiki/Anti-­pattern • You can also get a detailed list of the anti-pattern catalog at http://wiki.c2 .com/?AntiPatternsCatalog to learn more. • The concept of anti-patterns is not limited to object-oriented programming.

Q  &A Session 11.6 How are anti-patterns related to design patterns? Answer: When you use design patterns, you reuse the experiences of others who came before you. When you start blindly using those concepts for the sake of their use only, you fall into the trap of reuse of recurring solutions. This can lead you to a bad situation in the future, and then you identify that your return on investment (ROI) keeps decreasing but your maintenance cost keeps increasing. In simple words, the apparently easy and attractive solutions (or patterns) may cause more problems for you in the future. 11.7 A design pattern may turn into an anti-pattern. Is this understanding correct? Answer: Yes, if you apply a design pattern in the wrong context it can cause more trouble than the problem it solves, and eventually it will turn into an anti-pattern. So, before you start, ensure you understand the nature and context of the problem. For example, inappropriate use of the mediator pattern may end up with a God Class anti-pattern.

254

Chapter 11

Leftover Discussions

11.8 Anti-patterns are related to software developers only. Is this understanding correct? Answer: No. You have already seen various types of anti-patterns. So, the usefulness of an anti-pattern is not limited to developers; it may apply to others also. For example, it can be useful to managers and technical architects too. 11.9 Even if you do not get much benefit from anti-patterns now, they can help you to adapt new features easily with fewer maintenance costs in the future. Is this understanding correct? Answer: Yes. 11.10 What are the probable causes of anti-patterns? Answer: They can come from various sources or mindsets. A few common examples of what someone might say (or think) are listed here: •

“We need to deliver the product as soon as possible.”



“We have a very good relationship with the customer. So, at present, we do not need to analyze much about the future impact.”



“I am an expert of reuse. I know design patterns very well.”



“We will use the latest technologies and features to impress our customers. We do not need to care about legacy systems.”



“More complicated code will reflect my expertise in the subject.”

11.11 Can you mention some symptoms of anti-patterns? Answer: In object-oriented programming (OOP), the most common symptom is that your system cannot adopt a new feature easily. Also, maintenance cost is continuously increasing. You may also notice that you have lost the power of key object-oriented features like inheritance, polymorphism, etc. Apart from these, you may notice some or all of the following symptoms: •

Use of global variables



Code duplication 255

Chapter 11

Leftover Discussions



Limited/no reuse of code



One big class (God Class)



Presence of a large number of parameter-less methods, etc.

11.12 What is the remedy if you detect an anti-pattern? Answer: You may need to refactor your code and find a better solution. For example, here are some solutions to avoid the following anti-patterns:

256



Golden Hammer: You may try to educate Mr. X through some proper training.



Zero Means Null: You can use an additional Boolean variable that is more sensible to you to indicate the null value properly.



Management by Numbers: The numbers are good if you can use them wisely. You cannot judge the ability of a programmer only by the number of defects he/she fixes per week. Quality is also important. For example, fixing a simple UI layout is much easier compared to fix a critical memory leak in the system. Consider another example. “More tests are passing” does not indicate that your system is more stable unless these tests exercise different code paths/branches.



Shoot the Messenger: Welcome tester “John” and involve him immediately. Don’t consider him as your rival. You can properly analyze all of his findings and fix the real defects early to avoid last-­ moment surprises.



Copy and Paste Programming: Instead of searching for a quick solution, you can refactor your code. You can also make a commonplace to maintain the frequently used methods so as to avoid duplicates and have easier maintenance.



Architects Don’t Code: Involve architects in some parts of the implementation phase. It can help both the organization and themselves. This activity can give them a clearer picture of the true functionalities of the product. And truly, they should value your effort.

Chapter 11

Leftover Discussions

11.13 What do you mean by refactoring? Answer: In the coding world, the term “refactoring” means improving the design of existing code without changing the external behavior of the system/application. This process helps you to get more-readable code. At the same time, this code should be more adaptable to new requirements (or change requests) and they should be more maintainable.

Some Common Terminology It’s not the developer’s responsibility to only start with a good design. Maintaining a good design is equally important. If we focus on the quick fixes without maintaining the original design goals or architecture, we may encounter new issues. An inappropriate design can make an application rigid. Even if you start with a good design, the continuous quick fixes to this application can make it inefficient. Then a simple change may demand lots of effort. In the worst case, you see the fragility issue. What does this mean? In simple words: one small change in one location causes changes in multiple locations, and in the worst case, you discover that some of these areas are in no way related to the original change request. You can develop applications very quickly if you reuse some built-in parts that you or someone else developed earlier. When an entry-level programmer hears about reuse, he thinks that inheritance is his only available option, but this is not true. You have seen that in many situations object composition offers a better solution than inheritance. But the use of inheritance or composition becomes secondary if you bring in a code segment that is dependent on many other things, or already has potential errors; you are surely compromising quality. The inability to reuse software is often termed immobility. Viscosity is another important thing to consider in OOP. Wikipedia describes it as the ease with which a developer can add design-preserving code to a system. When maintaining the design, if you can add new code to your program easily, your program has a low viscosity. The opposite is obvious: in a high-viscosity design, adding hacks is easy, instead of working to preserve the original design. You can surely see that by using these hacks you make your system more rigid. This is one form of viscosity, which is also referred to as viscosity of design.

257

Chapter 11

Leftover Discussions

There is a different form called viscosity of environment. Consider a case where developers use a pillar build before they push a change in the main codebase. What do I mean by a pillar build? Let me give you an example: suppose your company developed a large application that has many components or modules. I refer to them as pillars. For example, if the application can send emails as well as faxes, we have an email pillar and a fax pillar. Since these components are big, to maintain separate components, the company needs separate teams. Each team can compile a specific pillar using a batch file to ensure that a new change in the pillar does not break other parts of the same ­pillar/component. This is a pillar build. So, you can think of it as a single-module build, or a single-component build. It is attractive when the full build (i.e., the complete compilation of all pillars) is a time-consuming activity, but you need to verify a critical fix as soon as possible. Let’s take an example: you have found a bug in the email pillar just prior to release your software. Let’s assume if you compile only the email pillar, it takes approximately 1 hour. But if you trigger a full build( which will compile email, fax, and other pillars together), it takes nearly 5 hours. So, to maintain your delivery schedule, you are tempted to run only an email pillar build. I believe that I do not need to tell you that relying on this kind of pillar build is risky if you work on interconnected modules. It is because unless you do not trigger the full build, you do not know whether the last-minute fix causes other pillars to break. Cohesion and coupling are two additional important concepts that were invented by Larry Constantine in the late 1960s. What do we mean by cohesion? The dictionary meaning of cohesion is interconnection or unity. In OOP, when you design a class, it measures the strength of the relationship between a class’s method and data. It will be easy for you if you can remember the single responsibility principle (SRP) from Chapter 4. These concepts go hand-in-hand, though cohesion is a more general concept. The opposite is coupling. Wikipedia says that coupling is the degree of interdependence between software modules. So, in OOP, you can say that it is a measure of interdependence between two classes. Let us say there are two separate classes, A and B. Now consider the case where A uses a B object in one of its methods, or you create a B class object inside the A class constructor and work on that. In these cases, A and B are tightly coupled. Even if B is a subclass of A and uses A’s method, you can say that they are tightly coupled. Remember that we want high cohesion and low coupling. I finish the chapter with Robert C. Martin’s words from “The Clean Code Blog” (which you can see at https://blog.cleancoder.com/uncle-­bob/2014/05/08/ SingleReponsibilityPrinciple.html): 258

Chapter 11

Leftover Discussions

“If you think about this (SRP) you’ll realize that this is just another way to define cohesion and coupling. We want to increase the cohesion between things that change for the same reasons, and we want to decrease the coupling between those things that change for different reasons.”

Q&A Session 11.14 You did not talk about the latest features in C# in this book. Is there any specific reason? Answer: We all know that change is the only “constant” in the software industry. Whatever is new today, can be outdated tomorrow. Yes, some new features are interesting. For example, in the latest edition of C#, you can execute a program without the Main() method using top-level statements. These statements execute in the order they appear in a file. For example, consider the following code: using System; namespace WithoutUsingTopLevelStatements {     class Program     {         static void Main(string[] args)         {             Console.WriteLine("Hello World!");         }     } } In earlier versions of C# and .NET, the namespace and the args parameter were optional. Now in .NET 5, you can have a further simplified version: using System; Console.WriteLine("Hello World!"); Remember that to compile this you have to use C# 9.0 (Target framework .NET 5.0); otherwise, you see the following: Error CS8400 Feature 'top-level statements' is not available in C# 8.0. 259

Chapter 11

Leftover Discussions

It is true that it’s a big simplification. This feature is useful for scripting scenarios. Now you type less and the code size is small, but you get the desired output without a problem. In this case, the .NET platform provides all the necessary things behind the scenes for you. Today, if a beginner starts with this shortcut, it can be hard for him to imagine the background stuff and the legacy code. (But it is possible that after few years, all beginners will prefer to start from here.) The truth is, when you join a corporate team, it is unlikely that every time you start off by using the latest features of a programming language. Instead, most of the time, you work with a legacy version to support existing customers. Unless the company decides you should work on an update, you will likely keep fixing the bugs in the legacy versions too. So, I always try to keep a balance. In my other books also, I use the fundamental features of a programming language so that you can understand them easily. I prefer to write codes that are supported in a wide range of versions. 11.15 I’d like to know whether you have used any other latest features in this book. Answer: Here are two more examples that you saw in this book: •

C# 9.0 allows you to check whether an object is not null. So, the following block of code does not create any compile-time error: if (sample is not null){ // some code }



From C# 8.0 onwards, you can use the following statement: using Sample sample = new Sample();

11.16 Can you suggest any general advice for me? Answer: I like to follow in the footsteps of my seniors and teachers who are experts in this field. And here are some general suggestions from them:

260



Program to a supertype (abstract class/interface), not an implementation.



Except for few cases, prefer composition over inheritance wherever you can.



Try to make a loosely coupled system.

Chapter 11

Leftover Discussions



Segregate the code that is likely to vary from the rest of your code.



Encapsulate what varies.

Summary Following the experts’ footprints and learning from recorded experience is a very good strategy. So, understanding design patterns is very important. At the same time, it is recommended that you use them wisely; otherwise, you might notice the impact of anti-patterns. As an obvious effect, you need to invest your time in refactoring the code or implementing a new design starting from scratch. At any cost, you should prefer an unattractive better fix over an attractive quick fix. I also describe some common terminologies at the end of this chapter. These often help you understand what a speaker says in a technical meeting or what a technical author writes in his book.

261

APPENDIX A

Winning Notes Congratulations! You have reached the end of the journey. Anyone can start a journey, but only a few will complete it with care. So, you are among the minority who possess the extraordinary capability to cover the distance successfully. I hope that you have enjoyed your learning experience, which can help you to learn and experiment further in this area. I said earlier that if you repeatedly think about the discussions, examples, implementations, and Q&A sessions in the book, you will have more clarity about them, you will feel more confident about them, and you will remake yourself in the programming world. Truly, a detailed discussion on any particular design pattern in depth would need many more pages, and the book would be too gigantic to digest. So, what is next? You should not forget the basic principle that learning is a continuous process. So, this book was an attempt to encourage you to learn the core concepts in depth so that you can continue learning in more depth. Still, I believe that learning and thinking by yourself will not be enough. I suggest you participate in open forums and join discussion groups to get more clarity on this subject. This process will not only help you; it will help others also.

A Personal Appeal to You Over the years, I have seen a general trend for my books. When you like the book, you send me messages, write nice emails, and motivate me with your kind words and suggestions. But most of these messages do not reach review platforms like Amazon and others. But when the opposite happens, I can see all the criticisms on those pages. I want to assure you that I know these criticisms help me to write better. But it will be helpful for me to know what you like about a book as well. These constructive suggestions can be included in an updated edition of the book.

© Vaskaran Sarcar 2021 V. Sarcar, Simple and Efficient Programming with C#, https://doi.org/10.1007/978-1-4842-7322-7

263

Appendix A

Winning Notes

So, I have a request for you: You can always point out the improvement areas of this work, but at the same time, please let me know what you liked about this book, and let others know about it in reviews. In general, it is always easy to criticize, but an artistic view and open mind are required to discover the true efforts that are associated with any kind of work. Thank you, and happy coding!

264

APPENDIX B

Resources This appendix lists some useful resources. A few of them are written by myself, and some of them use a different programming language. You can benefit from these books, or their updated editions: •

Design Patterns in C#: A Hands-on Guide with Real-world Examples by Vaskaran Sarcar (Apress, 2018)



Clean Architecture: A Craftsman’s Guide to Software Structure and Design by Robert C. Martin (Pearson, 2017)



The Pragmatic Programmer: From Journeyman to Master by Andrew Hunt and David Thomas (Addison-Wesley Professional, 1999)



Design Patterns: Elements of Reusable Object-Oriented Software by Erich Gamma et al. (Addison-Wesley, 1995)



Interactive C#: Fundamentals, Core Concepts, and Patterns by Vaskaran Sarcar (Apress, 2017)



Head First Design Patterns by Eric Freeman and Elisabeth Robson (O’Reilly, 2004)



Getting Started with Advanced C#: Upgrade Your Programming Skills by Vaskaran Sarcar (Apress, 2020)



The C# Player’s Guide (Third Edition) by RB Whitaker (Starbound Software, 2017)

The following are helpful online resources. These links are working fine at the time of this writing: •

https://dotnet.microsoft.com/learn/dotnet/what-is-dotnet



https://docs.microsoft.com/en-us/dotnet/csharp/programmingguide/classes-and-structs/destructors

© Vaskaran Sarcar 2021 V. Sarcar, Simple and Efficient Programming with C#, https://doi.org/10.1007/978-1-4842-7322-7

265

Appendix B

266

Resources



https://docs.microsoft.com/en-us/dotnet/standard/garbagecollection/fundamentals



https://docs.microsoft.com/en-us/dotnet/csharp/languagereference/language-specification/basic-concepts#automaticmemory-management



https://docs.microsoft.com/en-us/dotnet/csharp/whats-new/ csharp-9



https://docs.microsoft.com/en-us/archive/msdnmagazine/2005/july/discovering-the-design-patterns-you-realready-using-in-net



https://blog.cleancoder.com/uncle-bob/2014/05/08/ SingleReponsibilityPrinciple.html



https://devblogs.microsoft.com/dotnet/large-object-heapuncovered-from-an-old-msdn-article/



https://docs.microsoft.com/en-us/dotnet/standard/garbagecollection/unmanaged

Index A Abstract class, 17 AboutMe() method, 19, 20 analysis, 33 animal hierarchy, 18 class-specific implementations, 34 diamond problem, 20 Float()/Fly() method, 23 fundamentals, 17 inheritance hierarchy, 21–23 inheritances Boat class, 28 class diagram, 25, 28 code segment, 26 considerations, 27 hierarchies, 25 implementation/output, 29, 31, 32 vehicle/airplane/boat hierarchies, 28 interface, 22 SoftToys hierarchy, 18 Sound() method, 17, 18 supertype references, 24 Anti-patterns attractive solution, 250 classifications, 253 concepts/mindsets, 252, 253 design patterns, 254

history, 251 prevention, 251 refactoring, 256, 257 software industry, 251 symptoms, 255

B Behavioral patterns, 247–249

C Comments, 35 abstract class, 17 code segment, 36 documentation, 36 initial program analysis, 38 source code, 37, 38 multi-line comments, 35 Next method, 43 nutshell, 36 real-life scenario, 35 Rectangle class, 39, 40 SayHello() method, 42 single-line comments, 35 TODO comment, 40–42 Common Language Runtime (CLR), 204, 205, 208, 231 C# programming, 40–42

© Vaskaran Sarcar 2021 V. Sarcar, Simple and Efficient Programming with C#, https://doi.org/10.1007/978-1-4842-7322-7

267

Index

D, E Demonstration, 4, 7, 25, 37, 39, 54, 118 CommonLibrary.cs, 124–128 DIP, 100, 101, 103–105 DRY principle, 112–116, 118–121 Façades, 191, 192, 194–198 factories, 133–135, 138–146 garbage collector, 216–219 hook method, 173–176, 181–184 ISP, 93–97 LSP, 79, 84–87 memory leaks, 233–235 OCP, 61–65, 67–71 wrapper, 159–163 Dependency inversion principle (DIP) abstractions, 98, 103 analysis, 102 database parameter, 105, 106 demonstration, 100, 101, 103–105 high-level class, 100 low-level concrete class, 98 program hierarchy, 102 UserInterface and OracleDatabase, 99, 100 Design patterns behavioral patterns, 247 categories, 248, 249 class/object patterns, 249 creational process, 246 data structures, 250 Gang of Four (GoF), 245 histroy, 245 meaning, 244 PLoP, 245 programming languages, 250 structural patterns, 247

268

Diamond problem, 19–21, 152 Don’t repeat yourself (DRY) principle analysis, 114 BasicGameInfo reference, 123 code duplication, 109, 128 code segments, 110, 111 CommonLibrary.cs, 124–128 definition, 109 demonstration, 112–116, 118–121 DryDemoUsingDll project, 122 hard-coded strings, 114 initial version, 117, 118 instance methods, 114 key assumptions, 112 output process, 113, 121

F Façades assumptions, 188 benefits, 198 demonstration, 194–198 initial program analysis, 193 class diagram, 193, 194 demonstration, 191, 192 LoanStatus class, 190 optional parameters, 190 Person classes, 189 meaning, 187 problem statement, 188 program output, 197 Factories advantages, 147 analysis, 135, 136 AnimalFactory class, 135, 144 class diagram, 137

Index

demonstration, 133–135, 138–146 hierarchy, 136 modified implementation, 140 object-creation process, 131 output program, 135, 140, 143 problem statement, 132 program output, 147 requirements, 132, 141 stable application, 131 Finalizer vs. dispose analysis, 226 code segment, 221, 222 demonstration, 224–226 destructor, 227 Dispose() method, 223 Finalize() method, 228 IDisposable interface, 229 managed heap, 227 Microsoft documentation, 227 source code, 221 using statement, 230

G Garbage collection (GC), 204, 213 Garbage collector (GC) analysis, 219, 220 Collect() method, 215 demonstration, 216–219 descriptions, 214 different cases, 213 generations, 212, 213 GetTotalMemory() method, 214 memory leaks, 231 parameters’ descriptions, 215 phases, 213

H Hook method analysis, 176 demonstration, 173–176 enhanced requirement code segments, 180 demonstration, 181–184 GenerateGiftCoupon() method, 177–179 WashingMachine class, 179 initial program output, 170 PurchaseProduct class, 171, 172 PurchaseProduct() method, 172, 173 scenario, 170 SelectProduct() method, 172 problem statement, 169

I, J, K Interface, see Abstract class Interface segregation principle (ISP) analysis, 94, 97 class implementation, 88 demonstration, 93, 94 IFaxDevice/IPrinter interfaces, 95 implementation, 95–97 initial program class diagram, 89 fax hierarchy, 90 IPrinter class hierarchy, 89 PrintDocument()/SendFax() methods, 91 run-time error, 92, 93 SendFax() method, 91

269

Index

L Last In First Out (LIFO) mechanism, 205 Liskov Substitution Principle (LSP) analysis, 88 base class, 72 code segment, 73, 74 demonstration, 79–81, 84–87 IUser class, 82 LoadPreviousPaymentInfo() method, 77–79, 84 Main() method, 75, 82, 83 NotImplementedException, 83 output process, 76, 81, 88 ProcessNewPayment() method, 77, 84 Rectangle instance, 77 ShowArea() method, 73 Square class, 75–77 UserManagementHelper class, 78

M, N Martin, Robert C., 49, 58, 105, 107 Memory management finalizer/destructor analysis, 226 demonstration, 224–226 destructor, 227 Dispose() method, 223 Finalize() method, 228 IDisposable interface, 229 managed heap, 227 source code, 221 using statement, 230 garbage collection, 204 garbage collector, 203, 212–220 handling operations/unmanaged objects, 220, 221 IDisposable interface, 221 270

memory leak analysis, 203 definition, 231 demonstration, 233–235 diagnostic tools, 232, 235 garbage collector, 231 object count difference, 237 scenarios, 238 snapshots, 236–238 operating system (OS), 205 OutOfMemoryException, 204 stack/heap memory (see Stack/heap memory) Submit button, 203 symptoms, 203 Meyer, Bertrand, 58

O Object-oriented programming (OOP), 3, 47, 49, 250, 255 Open/closed principle (OCP) advantages, 71 analysis, 65 characteristics, 61 code segment, 60 definition, 58 demonstration, 61–65 EvaluateDistinction() method, 59, 65 IDistinctionDecider method, 66 inheritance, 58 initial program, 59–61 modified program, 67–71 program designing, 66, 67

P, Q Pattern Languages of Program Design (PLoP), 245 Polymorphism

Index

advantages, 14 analysis, 5–7 animal.Sound() method, 12 approach, 15 concrete implementation, 14 GetAnimal() method, 10, 11, 15 initial program, 3 Main() method, 8, 9 meaning, 3 running application, 11 Sound() method, 9 source code, 4, 5 subtype method, 12, 13 supertype, 14 version modification, 7, 8

R Return on investment (ROI), 250, 254

S Simple factory pattern, 135 Single responsibility principle (SRP), 50–51, 258 analysis, 54 data/properties/methods, 50 Employee class, 51 GenerateEmployeeId() method, 54 output, 53 responsibilities, 51 source code, 51–58 Software design patterns, 245 SOLID principles LSP, 72–88 DIP, 98–106 design principles, 49 fundamental guidelines, 106, 108

ISP, 88–98 OCP, 58–72 SRP, 50–58 Stack/heap memory allocate memory, 208 allocations, 208 C/C++ concept, 209 definition, 205 heap memory compaction technique, 212 garbage collector, 211, 214 scattered allocations, 210 unmanaged heap, 211 LIFO mechanism, 205 object reference, 207, 209 source code, 206 stages, 206 statuses, 207 Static/instance method AnimalFactory class, 241, 242 anti-patterns, 250–257 definition, 241 design patterns (see Design patterns) message, 242 ShowGreaterNumber() method, 243 terminology cohesion and coupling, 258 constant, 259 features, 260 general suggestions, 260 inappropriate design, 257 inheritance/immobility, 257 simplified version, 259 viscosity, 257

T, U Template method, see Hook method 271

Index

V Version management tool, 44

W, X, Y, Z Wrapper base classes, 152 BasicHome class, 153 definition, 146 interface, 152 object composition AddPlayGround() method, 158

272

analysis, 165 BufferedStream class, 165 class diagram, 158, 159 components, 153–158 demonstration, 159–163 FileStream class, 166 key concepts, 167 program output, 163 requirements, 156 SwimmingPool class, 158 problem statement, 150 subclassing, 150–153